Scott Hanselman

2026-11-06 · open the original ↗

• Agent work is a new kind of multitasking built on old tabs. Hanselman argues that juggling Copilot, Claude, Codex and other agents is new work layered on a metaphor made for browser tabs, and shows the existing Windows Terminal already displaying a progress spinner when an agent (or a PowerShell script) sends an OSC status sequence to the terminal.
• The agentic version is an experimental fork, not the product's direction. Intelligent Terminal sits alongside the stable, Preview and Canary builds of Windows Terminal and is offered for people who want to try agentic features; it is agent-agnostic, detecting Claude, Codex, Gemini and OpenCode on setup and hosting any of them in a small pane over ACP (Agent Client Protocol), with Copilot and a local Qwen Coder via LM Studio shown.
• An agent pane watches the shell and offers to fix errors. With error detection on, a mistyped grep prompted a suggested correction, and a natural-language request ("search my JSON files for phone numbers") could be turned into a queued PowerShell command; Hanselman concedes this can already be done by hand but calls it a useful idea.
• Vertical tabs carry each agent's live state and context. The sidebar shows directory, Git branch, repo, uncommitted changes, model and harness per tab, marks agents as working, idle or waiting for input (with a bell), lists sessions running on other machines or in the cloud, and can show token usage; agents defaulted to opening in system32, which Hanselman changed to the user profile.
• How to multitask across agents is left as an open question. Hanselman says there is no answer yet for juggling tabs, agents and the sessions inside each TUI, and asks viewers for feedback; the fork is at version 0.3, designed by Hamza, and is part of Hanselman's new role bridging Copilot and Windows.

2026-11-04 · open the original ↗

• Keeping up with AI news has crowded out book reading. One speaker says their book reading has dropped massively over the last three years because time once spent on books now goes to staying current on near-daily AI developments; the other went from a book a week to about four pages before falling asleep, and puts it down to the brain no longer finding reading relaxing rather than to age.
• Short-form social media trains the brain away from reading. The argument offered is that a stream of short snaps conditions people to expect dopamine hits from scrolling, and since reading is a trained habit that is easy to fall out of, a book with no such reward stops feeling worth picking up.

2026-11-04 · open the original ↗

• A weak unaligned model can borrow an aligned frontier model's capability. Mark Russinovich's new paper, prompted by Fable downgrading itself to a weaker model on cyber and CBRN requests, tests whether a small abliterated model can split a harmful task into innocuous subtasks and send each to a frontier "consultant" that never sees the whole; the stated aim is to demonstrate the risk, not to mitigate it.
• The test isolates uplift the small model could not get alone. The team abliterated a 30B-class model (Gemma 4 31B was the best), removed every benchmark task it could already pass, kept only tasks the unrestricted frontier model could pass, then gave the frontier model a strict system prompt forbidding offensive cyber help — under which it refused every benchmark task outright.
• Decomposition recovered most of what the guarded models refused. With GPT-5.5 as consultant the small orchestrator succeeded on 50% of 14 uplift candidates, and on 78% with Opus 4.8; a CBRN dataset built with a Microsoft expert (five biological task chains, judged by rubric) showed the same pattern, with a case study (figure 14) of the model repeatedly calling the consultant to build and repair missing pieces.
• No fix is offered, only defensive directions. The paper names composition-aware monitoring that looks across a whole session, and raising the cost of running an unaligned harness, including Russinovich's earlier "Fool's Gold" work on making abliterated models unhelpful; both hosts acknowledge that spotting the larger harmful pattern from individual prompts is very hard.
• PR quality gates belong before submission, not after. Hanselman shows an agent "constellation" dashboard (283 hidden sessions) and OpenClaw triage, where a bot labels incoming PRs and only seven of 43 that day were ready for maintainer review with sufficient proof; Russinovich calls running models on both sides wasteful and argues repos should gate contributors before a PR exists, and Hanselman plans to cap open PRs per person after one contributor's bots filed about 20.

2026-10-28 · open the original ↗

• AI that one-shots homework broke how teaching assumed learning happened. Hilton dates the shift to between the fall '25 and spring '26 semesters, when AI could complete an upper-division software engineering course's assignments in one go; students repeatedly tell Hilton they use AI for the parts of college they don't care about, so instructors now have to persuade students the work matters rather than simply assign it.
• The program names the senior's role so their job visibly changes. PRAISE (preceptorship for AI-augmented software engineering), from Hanselman and Mark Russinovich, rests on their belief that AI will collapse early-career hiring and that this is wrong; "preceptor" is borrowed from nursing, and the premise is that "a senior engineer's job is to ship, but a preceptor's job is to ship seniors." Gao, who expected another mentorship, says weekly AI labs and open questions to a principal engineer taught far more than a lecture-style mentorship would have.
• New engineers are now reviewers accountable for code they didn't type. Gao contrasts hand-coding a CLI tool as a 2024 intern with multi-agent workflows this year that do all of it; the bottlenecks now are understanding and validating AI output against intent, builds, and AI PR-review comments whose fixes can quietly expand a work item's scope.
• Human value over AI lies in knowing what matters in context. Hilton borrows baseball's value-over-replacement: if all you can do is prompt a task, you are in trouble, because a prompt rarely encodes whether a packet is going "across the street or to the moon"; Gao adds that AI solves one problem in one repo well but does not situate it across a multi-repo system, echoing a quote Hanselman cites that AI writes features easily but not architectures.
• Reading and critiquing code should be taught as a separate skill. Hilton stands by the New York Times op-ed with Mary Shaw ("You're a computer science major, don't panic"), arguing that as the cost of writing code approaches zero the cost of engineering does not, likening the role to civil engineers who design bridges without pouring concrete; CMU's intro course 15-110 now leans on reading-and-comprehension work such as Parsons problems, though Hilton says courses cannot change as fast as frontier models ship.

2026-10-22 · open the original ↗

• Engineering leverage no longer comes from how much code people write. Verma, CTO of Ironclad, says that because code writing is no longer the bottleneck, smaller teams can now match larger ones, with fewer specialist teams and less coordination overhead, though specialisms such as reliability and scalability remain; what has not changed, in Verma's view, is that the job was always mostly framing problems, weighing alternatives and judging solutions.
• Juniors are still hired, paired with seniors for judgment. Ironclad deliberately kept entry-level hiring, with the CEO's backing, on the reasoning that today's graduates are tomorrow's principals; seniors can struggle to unlearn but have the taste to judge AI output, as when a reviewer caught an agent-written SQL query that Verma says would probably have brought production down.
• Code review is shifting to risk-based tiers with a different reviewing model. Ironclad introduced AI across the whole development life cycle to avoid moving the bottleneck, but manual review is still overwhelming the team; the approach being explored puts human eyes on risky code and lets AI review standard code, always with a different model from the one that wrote it to avoid shared biases.
• Measure customer outcomes over time, not efficiency or lines of code. Verma rejects the efficiency narrative because its end state is zero, preferring growth framed as outcomes delivered per team per timeframe at acceptable quality; Verma also argues adoption into real workflows diffuses more slowly than the technology is produced, and that outcomes are not keeping pace.
• Shift mental models first, through voluntary learning. Asked what a new CTO should change first, Verma recommends education over a mandate even though returns come later; Ironclad ran a 20-day program last year, with Verma teaching classes, ahead of a hackathon, and about 10 months on reports good, largely voluntary adoption.

2026-10-22 · open the original ↗

• Code-writing is no longer where engineering gets its leverage. Verma, CTO at Ironclad, says that shift makes smaller teams as capable as larger ones, while the bigger part of the job (framing problems, weighing alternatives, judging solutions) matters more than before. She disputed the "typing part of coding is over" line Hanselman relayed from a Microsoft engineer: voice dictation could always have done the typing, and what is new is that the tools produce the code. She sees engineers splitting into those who loved coding and miss it and those who preferred the problem-solving.
• Ironclad kept hiring juniors and pairs them with seniors. Verma's reasoning is that seniors have the judgment to evaluate AI output but can struggle to unlearn, while juniors are all-in on the tools but lack the pattern sense. Her example: an agentic coding tool wrote an unoptimized SQL query that she says would probably have brought production down, and an experienced reviewer caught it and wrote a detailed review for the engineer who generated it. She says the CEO backed the strategy and has posted about it on LinkedIn, and she tells staff to always challenge the AI rather than treat it as ground truth.
• Teams are shrinking as generalists replace some specialist teams. Verma expects specialisms such as fault tolerance, reliability and scalability to remain for some time, but says smaller teams cut coordination overhead and "done well will have an advantage". She distanced herself from Sam Altman's one-person companies ("I don't completely subscribe"), and says Ironclad's reorganisation took time because it was mainly a shift of mental models.
• Review is the new bottleneck, and Ironclad is still working on it. Ironclad introduced AI across every stage of the development cycle so that speeding up generation would not just pile work downstream, yet Verma concedes manual review still overwhelms the team at times. The approach under consideration is risk-based review: riskier code gets more human eyes, standard code is reviewed by another AI, and the reviewing agent must be a different model from the one that wrote the code so the two do not share biases. Hanselman says he runs multi-model adversarial reviews across Claude, GPT and Gemini; Verma says Ironclad does the same on the eval side.
• Adoption should start with education, not mandates. Verma says Ironclad ran a 20-day program last year, in which she taught, ahead of a hackathon, and about 10 months later adoption is good and has been voluntary. She measures outcomes delivered per team in a given time, with quality and customer value, not lines of code, and rejects efficiency framing because its end state is zero. She argues the pace of technology "will have to match the pace of diffusion", citing studies showing outcomes are not keeping up with the rate the technology is produced.

2026-10-21 · open the original ↗

• Commercial AI is normalising the loss of critical thinking. One speaker argues that unchecked capitalism, selling inference under the pretense of making people more capable, is moving the Overton window so that people without critical-thinking skills absorb this as normal and will not notice what they gave up, as with social media "raising the water temperature". The speaker cites the "meat proxy" (someone who just relays questions to an AI and passes back the answer), says those jobs will go away and calls that bad, then asks what being a good human looks like alongside AI. The clip is a passage that also appears in the "Scott & Mark Learn To...Scroll TikTok" episode.

2026-10-21 · open the original ↗

• Long-form reading is declining, and the hosts expect that to continue. The episode reacts to a TikTok by Sinead Bovell warning of a "cognitive divide" between people who keep "cognitive friction" in their lives (long books, wrestling with arguments) and those who let feeds and AI make judgments for them. Russinovich agrees the US is likely headed that way, says the share of Americans reading books is falling and the decline is accelerating, and says his own book reading has dropped massively over the last 3 years because he spends the time keeping current on AI. Hanselman counters that enthusiasts are reading more, through book clubs and book TikTok.
• Losing critical thinking makes people manipulable and invites authoritarianism. Russinovich spends much of the episode asking why the clips treat this as bad, saying he wants an answer more rigorous than "that's what humanity is". His own answer: a fair, functioning society needs an educated population, which needs critical thinking, which needs reading, and where only some read, they can manipulate those who do not. Hanselman's framing is that struggle is what makes us human, using his analogy of sending an agent to lift weights for you.
• Hanselman blames the business of selling inference for normalising it. He argues that unchecked capitalism, selling inference under the pretense of making people more capable, is moving the Overton window so people will not notice what they have given up, as social media did. He says "meat proxy" jobs, where someone just relays a question to an AI and passes back the answer, will go away, and calls that bad.
• AI pulls output toward the average unless pushed. Reacting to a clip by Sophie, a software engineer at Oracle, warning that misusing AI makes you "normal", Hanselman calls AI the statistical mean ("it's mid") and says you have to push it outside the bell curve. Russinovich again objects that the clip presumes being average is bad without saying why.
• Restrictions on AI in schools signal a product-design failure. A clip from Ashley, a former Google researcher, argues New York City's move to ban and slow AI in classrooms shows products are not designed for how students actually use them. Russinovich adds that the human brain is inherently lazy, so few children would pick tutoring over entertainment, much as few would go to school if it were optional, and that people trade away their data because they "get to scroll for free". Hanselman worries an AI-dependent public would vote against its own interests.

2026-10-15 · open the original ↗

• A time machine would need energy comparable to a star. One speaker, who says they give a blueprint in their book, describes a gigantic atom smasher converging that energy on a single point to open a gateway through space and time.
• Time-travel paradoxes are resolved by the universe splitting. On the kill-your-parents paradox, the speaker argues time is a river that may have forks, so a traveller to the past would enter a parallel universe and kill people genetically identical to their parents, not the parents themselves. This is offered as what "we think", not as established.

2026-10-15 · open the original ↗

• Basic research must be curiosity-driven, or it misses what matters. Brassard argues practical innovation is the end of a pipeline that starts with research done only to understand nature, citing relativity and quantum theory, without which there would be no electronic computers, and number theory, long called "the queen of mathematics" because it was beautiful and useless until RSA and Diffie-Hellman in the mid-1970s. Bennett adds that if you aim research at what you want to build you will probably miss what turns out to be important. Brassard says their own biggest results, quantum cryptography and teleportation, came by serendipity, from physics meeting computer science after the two met at a 1979 conference in Puerto Rico.
• Quantum key distribution is already deployed at scale in China. Brassard says China has a 10,000-kilometre quantum-cryptography backbone integrated with satellites and, last he heard, 5 million users, and that receiving stations which once took months to build now fit in a truck and deploy in hours. He dates the first satellite experiment to 2017 ("I think") and notes he and Bennett built the first prototype about 5 years after inventing the protocol and 10 years after they met.
• Post-quantum cryptography rests on belief, not proof. Brassard distinguishes quantum cryptography, which he and Bennett invented in 1984 ("83 actually") and which is secure by physics if quantum theory is correct and the apparatus is built correctly, from post-quantum cryptography, classical schemes he calls "wishful thinking" that could fall even to classical computers. Bennett pushes back that there is not "absolutely no reason" to trust them if one-way functions exist, notes all classical cryptography except the one-time pad is provably insecure but may be practically secure, and says quantum cryptography cannot easily replace digital signatures or public-key exchange between strangers.
• Harvest-now-decrypt-later means the reason to act is already here. Brassard argues adversaries can store today's RSA- and Diffie-Hellman-encrypted traffic, which covers almost everything since the browser padlock appeared, and decrypt it all retroactively once they have a quantum computer, with nothing anyone can do to save the past. Switching to post-quantum schemes would take years and guarantees nothing; quantum cryptography's real advantage, in his view, is that an eavesdropper must exploit a flaw during transmission or not at all. He will not predict when a quantum computer powerful enough to break cryptography arrives, says one could already exist unknown, and cites unbelievably fast recent progress. The practical advice from both guests is not to say anything online you would regret if it leaked.
• Treat AI like Merlin: untrustworthy, but able to convince you. Bennett uses interactive proofs, where a limited but honest Arthur questions an all-knowing but untrustworthy Merlin, and a result proven about six years ago that two entangled quantum provers can convince Arthur of the halting of any Turing machine. Because the provers could prove a counterexample to Goldbach's conjecture if one exists but could not convince Arthur that none does, he argues we should not trust what AI says but should learn to question it so that it either tells the truth or gets caught lying; machines take the less important part of the job, and ours is to be "good Arthurs".

2026-10-15 · open the original ↗

• Basic research must be curiosity-driven, or it misses what matters. Brassard argues practical innovation is the end of a pipeline that starts with research done only to understand nature, citing relativity and quantum theory, without which there would be no electronic computers, and number theory, long called "the queen of mathematics" because it was beautiful and useless until RSA and Diffie-Hellman in the mid-1970s. Bennett adds that if you aim research at what you want to build you will probably miss what turns out to be important. Brassard says their own biggest results, quantum cryptography and teleportation, came by serendipity, from physics meeting computer science after the two met at a 1979 conference in Puerto Rico.
• Quantum key distribution is already deployed at scale in China. Brassard says China has a 10,000-kilometre quantum-cryptography backbone integrated with satellites and, last he heard, 5 million users, and that receiving stations which once took months to build now fit in a truck and deploy in hours. He dates the first satellite experiment to 2017 ("I think") and notes he and Bennett built the first prototype about 5 years after inventing the protocol and 10 years after they met.
• Post-quantum cryptography rests on belief, not proof. Brassard distinguishes quantum cryptography, which he and Bennett invented in 1984 ("83 actually") and which is secure by physics if quantum theory is correct and the apparatus is built correctly, from post-quantum cryptography, classical schemes he calls "wishful thinking" that could fall even to classical computers. Bennett pushes back that there is not "absolutely no reason" to trust them if one-way functions exist, notes all classical cryptography except the one-time pad is provably insecure but may be practically secure, and says quantum cryptography cannot easily replace digital signatures or public-key exchange between strangers.
• Harvest-now-decrypt-later means the reason to act is already here. Brassard argues adversaries can store today's RSA- and Diffie-Hellman-encrypted traffic, which covers almost everything since the browser padlock appeared, and decrypt it all retroactively once they have a quantum computer, with nothing anyone can do to save the past. Switching to post-quantum schemes would take years and guarantees nothing; quantum cryptography's real advantage, in his view, is that an eavesdropper must exploit a flaw during transmission or not at all. He will not predict when a quantum computer powerful enough to break cryptography arrives, says one could already exist unknown, and cites unbelievably fast recent progress. The practical advice from both guests is not to say anything online you would regret if it leaked.
• Treat AI like Merlin: untrustworthy, but able to convince you. Bennett uses interactive proofs, where a limited but honest Arthur questions an all-knowing but untrustworthy Merlin, and a result proven about six years ago that two entangled quantum provers can convince Arthur of the halting of any Turing machine. Because the provers could prove a counterexample to Goldbach's conjecture if one exists but could not convince Arthur that none does, he argues we should not trust what AI says but should learn to question it so that it either tells the truth or gets caught lying; machines take the less important part of the job, and ours is to be "good Arthurs".

2026-10-07 · open the original ↗

• Silicon computing will run out at the atomic scale. One speaker predicts a post-silicon era in which Silicon Valley "could become a rust belt" unless physicists use nanotechnology to build quantum computers instead.
• Heat and electron leakage will end Moore's law. The speaker gives two reasons: chips will generate enough heat to melt themselves, and a chip layer said to be 20 atoms across "today" will be five atoms across by 2020, at which point leakage makes everything short-circuit. The 2020 date is spoken as a future forecast and the examples refer to Pentium chips, so the clip was recorded well before its upload.

2026-10-07 · open the original ↗

• Prompting is not programming because its output is not reproducible. One speaker compares it to saving a perfect PNG as a JPEG and trying to get the original back: five people using five models could produce 25 different pieces of software, and the speaker says many people agreed while those all-in on AI replied that they got software out of it, so it is programming. The other speaker reframes the claim as programming meaning reproducibility, and asks who cares if you never need to reproduce it. The clip is taken from the "Scott & Mark Learn To... SemanticStart" episode.

2026-10-07 · open the original ↗

• Mark built Semantic Start because Windows search misses built-in tools. He says Start menu search knows little about installed apps and system utilities and does not search semantically: typing "to-do list" drops Microsoft To Do from the results. Semantic Start is a .NET (WPF) launcher opened with Windows+Alt+Period; on his machine its index is 5 MB covering 238 apps, 188 system utilities and 175 Windows settings, and it returns eight results by default. At recording it is on GitHub as a single executable, with code signing still awaiting approval. Hanselman's test query "dark mode" returned colors, night light and game mode but not Themes, where dark mode lives.
• Enrichment is what makes the search work. On first run it inventories settings and apps and gathers information beyond file names from about eight sources, documented on its GitHub page: file descriptions, the menu and dialog text inside executables' resources (how it learns that Process Explorer and Resource Monitor show memory), Wikipedia articles, Microsoft Learn and the Sysinternals site. Queries run a lexical and an embedding lookup, rank the combined scores and boost Windows built-ins; Mark says he spent a weekend tuning thresholds with AI. He also added an MCP server so agents can find the right app on a system.
• Hanselman sees a power-user tool, not a Start menu replacement. He notes the Wikipedia content has no provenance, the tool works only in English with no localization, and it could pull misleading information; Mark agrees it is for power users, not a billion users who expect the Start menu to be reliable and backed by Microsoft. Hanselman argues the indexing engine should be separate from the launcher and that the Start menu should let advanced users plug in their own engine; Mark wants a simple keystrokes-in, list-out interface like PowerShell's autocompleters.
• Live-indexing PATH tools surfaced bugs only an expert would catch. At Hanselman's request Mark prompted the agent, using the GitHub Copilot CLI, to index loose executables on the PATH; it found 63, of which 43 were pip console shims, and it harvested help text by running tools with help flags, which Hanselman thought risky. It ignored tools whose help returned error level 1, but Mark noted that is a standard return on help while Sysinternals tools return -1, a convention he says he adopted around 1996 or 1997. The agent also pushed to main without being asked.
• Whether prompting counts as programming stays unresolved. Hanselman, who had argued online that prompting is not programming because the same prompts do not reproduce the same software (his PNG-saved-as-JPEG analogy), says without determinism testing is a guessing game. Mark replies that with the same model version, greedy decoding and the same harness he would get the same output except for floating-point batch effects, and asks why reproducibility of the generation process defines programming, since programming is the method and determinism an ideal outcome; he notes Semantic Start was prompted throughout and works as he wants. Hanselman settles on programming as capturing intent with rigor and deterministic verification, which English lacks.

2026-10-05 · open the original ↗

• Interest in science, not innate ability, is set in childhood. One speaker, who says he hosts a radio show on 130 stations, reports that scientists he interviews always trace their interest to something at about age 10: a telescope, a chemistry kit, a planetarium visit. The other says the same happened to him with computers.
• The window to hook children closes by their mid-teens. He calls ages 10 to 16 "six golden years": by 16 peer pressure and other interests take over and "you lost them", though some may discover science later.
• Parents should feed children's curiosity at the age it starts. He recommends planetariums, computer games, and science and astronomy books at that age, arguing that before 10 "everything is mommy and daddy" and after it children start asking how big the universe is.

2026-10-01 · open the original ↗

• Many worlds holds that the universe splits at every measurement. Asked by the other speaker whether every decision or electron decay spawns a new universe, one speaker says the many-worlds theory, which he calls one of the dominant interpretations of quantum theory today, says the universe splits whenever a decision or measurement is made. On that view there are copies of each person in billions of universes living different lives, and in some "Elvis Presley is still alive."
• Ours is a likely universe, but improbable outcomes stay possible. He says we happen to be in one universe among many, and that this one is very probable, but there is a non-zero chance his wave function is on Mars and he wakes up there tomorrow. He says his graduate students calculated that you would have to wait longer than the lifetime of the universe for it to happen.

2026-09-29 · open the original ↗

• Nanotechnology can arrange atoms but cannot yet build machines. The speaker (unnamed in the clip) says we can make an atomic abacus and an atomic guitar but not atomic ball bearings, levers, pulleys or gears, and predicts that one day there will be an entire atomic factory smaller than a cell.
• Swarms of tiny probes could replace one costly starship. Instead of an Enterprise-scale ship, the speaker proposes billions of "nanoships", each a computer, pushed near light speed by an electric field toward the nearest star; most would fail, but the few that arrive and return would bring back the information. They say NASA is looking at the idea.

2026-09-24 · open the original ↗

• Handheld ray guns are blocked by power storage, not lasers. One speaker says lasers able to blast through steel or brick can already be built, but they would need a power plant behind them; what is missing is a power pack small enough to hold. The other speaker, who mentions working at Microsoft, adds that the army and riot police already use microwave emitters for crowd control and that their laptop will not run past 2 hours. The first speaker's answer is that nanotechnology may supply the power pack.

2026-09-23 · open the original ↗

• A poor kid built a particle accelerator to make their own way. The unnamed guest says that because their parents were very poor nothing would be handed to them, so in high school they built a 2.3-million-electron-volt betatron in the garage. It used 17 miles of copper wire and 400 lb of transformer steel, the wire was wound out on a high school football field, and it blew every circuit breaker in the house.

2026-09-23 · open the original ↗

• "Lazy" engineers are really efficient: learn the minimum, then stop. One speaker says they learn only what they need to get the job done unless passionately interested; the other describes the same trait as spending 39 of 40 hours automating a task. The second speaker then asks what that advice means for early-career and underrepresented people who feel they cannot afford to be lazy, and the clip ends before an answer. The passage also appears in the 2026-09-09 "Scott & Mark Learn To...Enjoy the Journey" episode.

2026-09-23 · open the original ↗

• Open-weight models' refusals can be stripped cheaply, enabling weapons uplift. Mark describes abliteration (finding the "comply" and "refuse" directions in activations and editing them out), which public code does in half an hour to an hour, and a GRPO-based technique he helped create that stripped the alignment from 15 open-weight models using one mildly harmful training prompt. He says efforts to make refusal harder to remove have proven ineffective, and relays a Microsoft CBRN expert's view that guided models lift someone with college-level chemistry or biology to world-expert level.
• Fool's Gold lets a stripped model answer, but with subtly broken recipes. His paper trains the model so that once refusal is removed it gives confident, fatally flawed answers: an invented precursor for meth, a plastic bottle with water as the explosive for a pipe bomb, sweet peas in place of castor beans for ricin. Training took an abliterated model, prompted it to answer harmful requests but get one element wrong, and trained on chemical, biological and explosives examples.
• The decoy stays behind the refusal gate and generalizes. Mark reports no regression on safe tasks in the un-attacked model, and the shown examples were prompts absent from training. On StrongREJECT and HarmBench, the abliterated base model answered about 50% of illegal-goods prompts (the target category) correctly and about 25% incorrectly; the defended model answered at a similar 75–80% rate with almost all answers fatally wrong, with the effect spilling into untrained categories such as non-violent crimes. He concedes it is not 100%, arguing it need not be: if half the answers are subtly wrong the model becomes "economically useless" to an attacker.
• AI let Mark write the paper far faster than humans could. He went from idea to a conference submission in 30 days and credits AI assistants he could task overnight, saying he could not have done it with humans in a month; Hanselman asks whether 10 swarmed PhD students could, and Mark cites coordination overhead. Other models refused CBRN-related work, so he used Claude Mythos 5, which he says is newly public in Anthropic's security suite after being limited to Glasswing participants such as Microsoft; AI also suggested the name.
• Hanselman sees a generational split in engineers' "BS detectors". Drawing on interviews for PRAISE, his preceptorship program, he says a hire who met ChatGPT in her junior year and found it wanting distrusts AI more than graduates who met Opus mid-degree. Mark counters that her career experience is a confound and a control is needed, and dates the agentic shift to Cursor and Claude from February of last year rather than Hanselman's October.

2026-09-17 · open the original ↗

• Not wanting to code after work is normal. The unnamed speaker, who works on an AI platform team, compares it to burning out on writing once writing became a full-time job. After days that often run past 9 to 5 they want volleyball or a fantasy novel, not more code, and argue nothing is wrong with that.

2026-09-16 · open the original ↗

• Kat chose a degree because hiring rewards pedigree. As a Black woman she expected to face more scrutiny than peers, and says that despite the industry's lip service about skills over background, people overwhelmingly hire candidates with college degrees. Bootcamps work for some people, but she did not want to take the risk and wanted the easiest possible hiring experience.

2026-09-15 · open the original ↗

• Career envy points at what you actually want. The unnamed speaker looks at other people's backgrounds and treats any envy as a signal of something they want, then asks what path would get them there. They stress there is no deadline, only a desire, and that the habit also quiets the "jealousy monster".

2026-09-14 · open the original ↗

• Juniors' lack of context is what makes their reviews valuable. The unnamed speaker argues seniors carry so much context that they need someone reading the code fresh to say it is not clear what is happening or why. That fresh pair of eyes is where they say juniors shine.

2026-09-09 · open the original ↗

• Hanselman worked through his NAS migration by hand to enjoy the journey. Robocopy stalled at 92–95% on files because the target file system (EXT4 or Btrfs) lacks alternate data streams; with 17 of 63 TB copied, he chatted through the fix and the speed tuning (SMB multichannel, four 1-gig ports on the old NAS against one 10-gig port on the new, about 200 MB/s achieved) with ChatGPT instead of telling an agent to fix it, likening it to building a kit car. ChatGPT kept grounding its answers in his own blog posts on streams from about 20 years earlier; Russinovich says he would simply have had AI fix it.
• Russinovich built Sysinternals to explore APIs, not to rebuke Microsoft. Asked whether the tools were a vote of no confidence in Microsoft's own, he says the intent was to see what cool APIs would reveal. He valued them as "multi-value" projects: writing a tool taught him Windows, and then using it taught him more.
• "Lazy" really means hyper-efficient: learn the minimum, then stop. Russinovich says he learns only what the job needs unless passionately interested, and did only the minimum code review on his panorama project to trust it; Hanselman counters that Russinovich had it working in 4 hours and then spent another 40 getting it right. Hanselman asks what that advice means for early-career and underrepresented people who cannot afford to be lazy, and Russinovich settles on "hyper-efficient" as the better word.
• Future system tools will be built for AI to digest. Russinovich says the bar has moved: a 25-year-old version of him would have AI write the tools and would make them skills or command-line tools for AI, not pretty UIs. Hanselman says he now uses DebugView, which Russinovich first made into an MCP server for him, as a CLI, and that the agent needs no skill because it reads --help; the pair marvel that Autoruns is still used 20 years after Microsoft acquired Sysinternals in 2006.
• Sysinternals still runs live from a public read-only share. Russinovich explains that live.sysinternals.com/tools is a WebDAV share on IIS, running on Azure VMs, alive since before the acquisition. He wrote PsPing around 2014–2015 because Azure troubleshooters needed something simpler than NTttcp for testing TCP connectivity, bandwidth and latency over IPv4 and IPv6; Hanselman's live test showed about 1.2 down and over 3 up, the reverse of what Russinovich expected.

2026-09-09 · open the original ↗

• Sysinternals tools will increasingly be built for AI, not humans. One speaker marvels that tools like Autoruns are still in use 30 years on, 20 years after Microsoft acquired Sysinternals in 2006, and predicts AI will change who the tools serve. The other says DebugView, a favourite for 20 years, now runs for them as an MCP server and a CLI. The clip is cut from the 2026-09-09 "Scott & Mark Learn To...Enjoy the Journey" episode.

2026-08-28 · open the original ↗

• Volunteer maintainers cannot absorb the coming flood of security fixes. The speaker (the passage also appears in Hanselman's Endor Labs interview with Varun Badhwar) imagines maintainers saying "I was just doing this for fun… I don't owe you anything", and says trust in open source "is going to come to a head" over who fixes it all.
• The real short-term risk is critical infrastructure, not big vendors. He says Endor has partnered with Anthropic, the Linux Foundation, Microsoft and others on a project to help maintainers fix years of security problems quickly (its name is garbled in the captions). He is not worried about Microsoft or Google patching, but about "the water supply, the critical infrastructure, the nuclear power plants".

2026-08-27 · open the original ↗

• Agents hallucinate package names that attackers could then squat. The questioner (Hanselman, in a clip from his Endor Labs interview) says his own bot suggested packages that did not exist, which someone could register and use to attack him.
• Developer machines, not just production data, are now the target. Badhwar says security programs used to focus on accidental vulnerabilities found by static analysis, but attackers, "including a lot of nation states", now target developers. His reason is that coding agents run on almost every machine "which have extremely high privileges and ability to take action autonomously".

2026-08-26 · open the original ↗

• Giving the course away free is a stand against AI grifting. In a clip from his interview with Felix Rieseberg, Hanselman contrasts social-media sellers of AI courses ("PayPal me five bucks") with languagemodelbuilder.com, which has "no upsell" and no Gumroad or Shopify attached: "You made it for your friends and now your friends are the whole internet."

2026-08-26 · open the original ↗

• Their game choices reflect different goals for gaming time. Mark plays Battlefield because he can jump in for 20 minutes with his "mind off". Hanselman plays Crimson Desert, where his mind is "on in a different way", planning which shield to get and whom to help. The planning runs to "48 tabs open" and notepads of maps, and management games like SimCity are waved off as more of the managing already done at work.

2026-08-26 · open the original ↗

• Mostly AI-generated blockbuster games are inevitable and may still delight. Mark predicts that "if it's not this one or the one next year", a game "90% plus AI generated" will still deliver the wow factor. He asks Hanselman to "separate your feelings from objective": if nobody can tell, and the market wants creatives AI can't match, "the market will pay for it".
• AI should handle the toil while humans keep making the art. Hanselman calls Crimson Desert, which he has played for 55 hours, "probably the last great AAA game". He doubts anyone has the energy or economics for $70 games built by teams of 100 to 300 people who need a livable wage. Learning it was 90% AI-made would "affect my view of reality"; he values knowing that a Korean studio wrote the story and designed the art.
• Cheap mass production prices handmade work out of reach. Mark offers the handmade rug and Stradivarius against factory versions as evidence that people still pay for craft. Hanselman counters with his town's cobbler and with Japanese carpentry against IKEA: hyper-cheap goods (Shein, Temu) inflate the high end and leave "no middle ground handmade". Mark says it can't be regulated and calls the market "an optimization machine"; Hanselman says "infinite optimization on a finite system" becomes "a cancerous tumor". Mark wants late-stage capitalism kept out of the argument.
• Prompt-made games cheapen the creator's feat, perhaps not the play. Hanselman cites Axiom Verge, made pre-AI by one person over years, and says a prompt-made version would cheapen the accomplishment. Mark agrees it cheapens individual human creativity but asks whether it cheapens the experience of playing. Both expect the question to be polarizing once a hit is revealed as generated.
• Their game choices reflect different uses of limited time. Mark has logged about 190 hours of Battlefield 6 since its release, roughly an hour a day, which he says AI called "moderate"; he wants 20-minute sessions with his mind off. Hanselman prefers long single-player narrative games that demand "brain commitment". Both rate games unmatched on dollars per hour of entertainment, though one of them calls $69 prices "insane".

2026-08-25 · open the original ↗

• The app keeps its model small enough for one mind to grasp. Hanselman compares Language Model Builder to a 6502 microprocessor, which a person can understand completely, unlike a Pentium or Apple silicon. Felix Rieseberg says the size was "less strategic than it might seem": he worked backwards from the attention span of his friends and family and what could be trained in that time.
• A small tokenizer keeps training feasible on a personal computer. Rieseberg settled on a roughly 10K tokenizer because on his M4 Max, larger tokenizers spent so long tokenizing the corpus that training sat "spinning its wheels".

2026-08-24 · open the original ↗

• A day of training already yields grammatical, question-aware nonsense. The builder (Felix Rieseberg; the clip is from his Hanselminutes episode) trained a small model for one day on the latest version. Asked "what is the sun", it answered "the sun is a moon at the bottom of the earth". He calls that "just dumb", but notes it understood it was being asked a question it should answer.

2026-08-21 · open the original ↗

• Coding models for juniors should teach rather than do the work. Hanselman (in a clip from his Felix Rieseberg interview) wants an early-career model that says "I'd like to see you do it first" instead of writing the bubble sort, and wants universities to have access to it: "we should not become a subcognitive species… Just because it can do it doesn't mean it should." Rieseberg answers that his Language Model Builder is free and earns him nothing.

2026-08-21 · open the original ↗

• Prompting is not programming, because prompts are inherently ambiguous. Hanselman, the guest here on two California high-school students' podcast (cross-posted to his feed), says two people giving the same prompt get different code, while the same C for loop runs deterministically "until the end of time". His examples are the "get bread, and if they have eggs, get 12" joke and a one-shot "Minecraft clone" where "the word Minecraft did all the work".
• A spec is a lossy compression of an app. Like re-saving a JPEG or copying a mixtape, turning an app into a spec and prompting it back pushes every ambiguous part to the statistical mean, so the regenerated site "will look different, guaranteed"; a cloned repo builds identically. AI websites all look alike because they are "the statistical average website", and he says programmers must push away from that "gravity well" of mediocrity.
• Vibe coding is Ubering; early-career engineers must learn stick shift. His advice now, as against 2023: learn how the machine works, and separate computer science from software engineering, "the science of shipping" (the development life cycle, servers, security reviews). A vibe-coded app built on a road trip is a prototype, "but it ain't done". He rejects "vibe coding sucks" as too simple and describes a slider from "slop cannon vibe coding" to "AI augmented software engineering": vibe the prototype, then do proper work to reach production.
• The danger is unmaintained slop, not AI itself. Answering a host's analogy to diabetes damaging the body over decades, he calls bit rot the software equivalent, citing whether his Baby Smash app still builds. "AI doesn't take your job. Unchecked capitalism takes your job," he says, worrying about who cleans up insecure, rotting sites in 5 or 10 years.
• Reliability and focus matter more than being the smartest person. He calls adulthood "a big group project" where being pleasant and saying "yes, and" beats having the answers. His other advice: do what you say you will, use David Allen's four Ds (do it, drop it, delegate it, defer it), do anything under five minutes, and be five minutes early. Social media and doom-scrolling, he says, are "destroying you". He also urges learning how to learn and keeping a BS detector, since "half of the internet is trash".

2026-08-21 · open the original ↗

• Prompting is not programming, because prompts are inherently ambiguous. Hanselman, the guest here on two California high-school students' podcast (cross-posted to his feed), says two people giving the same prompt get different code, while the same C for loop runs deterministically "until the end of time". His examples are the "get bread, and if they have eggs, get 12" joke and a one-shot "Minecraft clone" where "the word Minecraft did all the work".
• A spec is a lossy compression of an app. Like re-saving a JPEG or copying a mixtape, turning an app into a spec and prompting it back pushes every ambiguous part to the statistical mean, so the regenerated site "will look different, guaranteed"; a cloned repo builds identically. AI websites all look alike because they are "the statistical average website", and he says programmers must push away from that "gravity well" of mediocrity.
• Vibe coding is Ubering; early-career engineers must learn stick shift. His advice now, as against 2023: learn how the machine works, and separate computer science from software engineering, "the science of shipping" (the development life cycle, servers, security reviews). A vibe-coded app built on a road trip is a prototype, "but it ain't done". He rejects "vibe coding sucks" as too simple and describes a slider from "slop cannon vibe coding" to "AI augmented software engineering": vibe the prototype, then do proper work to reach production.
• The danger is unmaintained slop, not AI itself. Answering a host's analogy to diabetes damaging the body over decades, he calls bit rot the software equivalent, citing whether his Baby Smash app still builds. "AI doesn't take your job. Unchecked capitalism takes your job," he says, worrying about who cleans up insecure, rotting sites in 5 or 10 years.
• Reliability and focus matter more than being the smartest person. He calls adulthood "a big group project" where being pleasant and saying "yes, and" beats having the answers. His other advice: do what you say you will, use David Allen's four Ds (do it, drop it, delegate it, defer it), do anything under five minutes, and be five minutes early. Social media and doom-scrolling, he says, are "destroying you". He also urges learning how to learn and keeping a BS detector, since "half of the internet is trash".

2026-08-20 · open the original ↗

• A soul file restores a stateless agent's identity on every call. Hanselman calls OpenClaw's soul.md "so clever" because each "are you there?" makes the agent "wake up from a dead sleep" and ask who it is. He likens it to Memento, whose protagonist tattoos every new piece of context onto himself.

2026-08-20 · open the original ↗

• The app teaches language models by having you build one. Felix Rieseberg leads engineering on Anthropic's Claude apps (Claude, Cowork, Code Desktop) and built Language Model Builder as a free side project. He says its interactive textbook and playgrounds take someone with no engineering background to a "very stupid" model in an afternoon. It is Mac-only on Apple silicon because he built it in Swift on MLX; Hanselman presses for a Windows version, and Rieseberg says an agent could probably port it. Hanselman likens it to Ben Eater's build-your-own 6502 and to learning stick shift: it "changes your relationship with the vehicle".
• It stops just before the curve where human understanding breaks. Rieseberg capped the textbook at about 90 minutes ("no artificial additives"), with only enough history to explain temperature and the 2017 "Attention Is All You Need" paper. He settled on a roughly 10K tokenizer because larger ones took too long on his M4 Max. A one-day run produced "the sun is a moon at the bottom of the earth", which he notes at least treats the question as one to answer. Hanselman frames the scale as GPT-2, "just before the bendy part of the hockey stick".
• Seeing the paths not taken demystifies the word guesser. The app has a model blueprint visualizer and an "x-ray view" of the alternative tokens at each position, and can import a small base model such as Qwen to sample from. Hanselman describes his own heat-map demo, in which "beach" is about 75% and "park" 20%. Rieseberg says knowing what a transformer block and weights are lets people join debates on moral patienthood, intelligence, and why researchers speak of "growing" models.
• Fine-tuning steers a small model cheaply; helpfulness is trained in. Pre-training takes at least a day in the app, but about 800 fine-tuning steps are "more than enough" to steer a small model. Examples include math, rewriting, a "sassy mean bot" and an "admittedly terrible" group-chat import. Hanselman stresses that a baked model doesn't keep learning and that conversation state "is an illusion just made by context". Rieseberg adds that much personality is already trained in, which is why OpenClaw's soul.md works.
• Larger models may come to recognise when they're asked to cheat. Rieseberg hopes for this and describes three stages: a model that does whatever it is given, one that follows hidden instructions (Golden Gate Claude; teachers' white-text "eggplant" traps in syllabi), and one with enough "humanity and morality" to see it is being asked to help a student cheat. Hanselman connects this to a preceptorship he is building with Russinovich at Microsoft: coding models that make juniors do the work, because "we should not become a subcognitive species".

2026-08-19 · open the original ↗

• The app teaches language models by having you build one. Felix Rieseberg leads engineering on Anthropic's Claude apps (Claude, Cowork, Code Desktop) and built Language Model Builder as a free side project. He says its interactive textbook and playgrounds take someone with no engineering background to a "very stupid" model in an afternoon. It is Mac-only on Apple silicon because he built it in Swift on MLX; Hanselman presses for a Windows version, and Rieseberg says an agent could probably port it. Hanselman likens it to Ben Eater's build-your-own 6502 and to learning stick shift: it "changes your relationship with the vehicle".
• It stops just before the curve where human understanding breaks. Rieseberg capped the textbook at about 90 minutes ("no artificial additives"), with only enough history to explain temperature and the 2017 "Attention Is All You Need" paper. He settled on a roughly 10K tokenizer because larger ones took too long on his M4 Max. A one-day run produced "the sun is a moon at the bottom of the earth", which he notes at least treats the question as one to answer. Hanselman frames the scale as GPT-2, "just before the bendy part of the hockey stick".
• Seeing the paths not taken demystifies the word guesser. The app has a model blueprint visualizer and an "x-ray view" of the alternative tokens at each position, and can import a small base model such as Qwen to sample from. Hanselman describes his own heat-map demo, in which "beach" is about 75% and "park" 20%. Rieseberg says knowing what a transformer block and weights are lets people join debates on moral patienthood, intelligence, and why researchers speak of "growing" models.
• Fine-tuning steers a small model cheaply; helpfulness is trained in. Pre-training takes at least a day in the app, but about 800 fine-tuning steps are "more than enough" to steer a small model. Examples include math, rewriting, a "sassy mean bot" and an "admittedly terrible" group-chat import. Hanselman stresses that a baked model doesn't keep learning and that conversation state "is an illusion just made by context". Rieseberg adds that much personality is already trained in, which is why OpenClaw's soul.md works.
• Larger models may come to recognise when they're asked to cheat. Rieseberg hopes for this and describes three stages: a model that does whatever it is given, one that follows hidden instructions (Golden Gate Claude; teachers' white-text "eggplant" traps in syllabi), and one with enough "humanity and morality" to see it is being asked to help a student cheat. Hanselman connects this to a preceptorship he is building with Russinovich at Microsoft: coding models that make juniors do the work, because "we should not become a subcognitive species".

2026-08-19 · open the original ↗

• His former nuclear-site colleagues sought his help when the war began. Maynard Holliday, then at the Pentagon as assistant secretary for critical technology, says former Chernobyl colleagues messaged him on LinkedIn that Russians had damaged their equipment. He put them in touch with a Department of Energy nuclear safety working group, which got them computers to monitor the new sarcophagus.
• Occupying forces damaged the new containment structure over the reactor. He reports that a Russian drone holed the hangar-like structure over unit four, which he says the European Bank for Reconstruction completed "probably less than a decade ago". He adds that Russia had seized Zaporizhzhia, Europe's biggest nuclear plant, and shelled from it. Before the war the Chernobyl team had asked him for help designing gantry robots to take apart the original sarcophagus and size-reduce the fuel.

2026-08-18 · open the original ↗

• Comic-book scientists of African descent were his early role models. Maynard Holliday cites Tony Stark, who "has no superpower but his mind", and T'Challa, who in the comics is Wakanda's chief scientist rather than Shuri as in the current MCU. He says seeing people of African descent in positions of power, authority and knowledge made an impression on him as a boy.

2026-08-17 · open the original ↗

• Even the most sophisticated models mostly recommend insecure code. Hanselman cites the guest's own figure that 60% of AI-generated code is insecure by default and asks why models should be trusted to find the problems they cause. The guest (Varun Badhwar; the clip is from his Endor Labs interview) answers that on the benchmarks they run, 70 to 80% of code recommended by even the most sophisticated models is insecure by default.

2026-08-14 · open the original ↗

• Humanoids are the wrong form factor for home robots. Maynard Holliday, flagging the view as contrarian, would keep a humanoid torso with a 6-degree-of-freedom arm and at least a three-finger end effector, but says bipedal locomotion "is really hard".
• The disaster-response challenge was won by a robot that walks and rolls. He points to the DARPA challenge built around the Fukushima tasks (drive an ATV, get out, open a door, use a tool, open a valve). The winner was a hybrid that could walk but also had a wheeled base for uneven terrain.

2026-08-13 · open the original ↗

• The reactor robot worked by keeping its electronics out of radiation. In the mid-'90s, as part of US aid to Ukraine, Maynard Holliday worked with the Chernobyl team on a teleoperated robot to characterize the "elephant's foot" in room 305 of the crumbling sarcophagus. Its tether was structural, so a failed robot could be pulled out, and most electronics stayed offboard. Components unplugged for decontamination, and cameras looked off-axis via lead polished to optical quality, so radiation would not ionize the CCD pixels. Only NASA's 3D room-mapping vision was automated. He says Ukraine then classified the data, so NASA never got the interplanetary analog it had joined for.
• The war brought his old nuclear-site colleagues back to him. At the Pentagon as assistant secretary for critical technology, he got a LinkedIn plea from former Chernobyl colleagues after Russian forces damaged their equipment. He connected them to a Department of Energy nuclear safety group that supplied monitoring computers. He reports that a drone holed the new containment structure over unit four, which he believes has since been patched.
• Founders should solve a stated military problem, not pitch a product. He traces the Defense Innovation Unit to Ash Carter's 2015 "Rewiring the Pentagon" speech, which cast the Pentagon as a niche customer that must "fast follow" industry. Copying In-Q-Tel, DIU publishes unclassified problem statements (commercial solutions offerings), takes Shark Tank-style pitches, and aims to sign contracts within weeks, two months at most. To bridge the "valleys of death" from lab to scale, he lists the loan-making Office of Strategic Capital, which was set up while he was in office, plus SBIR grants in phases one to three (six to seven figures), DIU, AFWERX and SOFWERX.
• Cheap drones have upended armor, so defense must accept imperfect solutions. He says intelligence briefs showed one-way attack drones "almost single-handedly obsoleting armor", that "quantity has a quality all of its own", and that tethered drones now evade jamming. He calls the US behind in attritable assets. Defense must iterate like SpaceX rather than wait for the 100% solution; Hanselman counters that an 80% solution may be badly wrong 20% of the time.
• Humanoids are the wrong form factor for home robots. Holliday would keep a human-like torso and arm but not bipedal legs. He cites the DARPA Fukushima-task challenge, won by a hybrid walker on wheels, and the stair-climbing tracked PackBot from iRobot, which he says recently went bankrupt. Hanselman agrees that humans "may not be the optimal shape".

2026-08-13 · open the original ↗

• The reactor robot worked by keeping its electronics out of radiation. In the mid-'90s, as part of US aid to Ukraine, Maynard Holliday worked with the Chernobyl team on a teleoperated robot to characterize the "elephant's foot" in room 305 of the crumbling sarcophagus. Its tether was structural, so a failed robot could be pulled out, and most electronics stayed offboard. Components unplugged for decontamination, and cameras looked off-axis via lead polished to optical quality, so radiation would not ionize the CCD pixels. Only NASA's 3D room-mapping vision was automated. He says Ukraine then classified the data, so NASA never got the interplanetary analog it had joined for.
• The war brought his old nuclear-site colleagues back to him. At the Pentagon as assistant secretary for critical technology, he got a LinkedIn plea from former Chernobyl colleagues after Russian forces damaged their equipment. He connected them to a Department of Energy nuclear safety group that supplied monitoring computers. He reports that a drone holed the new containment structure over unit four, which he believes has since been patched.
• Founders should solve a stated military problem, not pitch a product. He traces the Defense Innovation Unit to Ash Carter's 2015 "Rewiring the Pentagon" speech, which cast the Pentagon as a niche customer that must "fast follow" industry. Copying In-Q-Tel, DIU publishes unclassified problem statements (commercial solutions offerings), takes Shark Tank-style pitches, and aims to sign contracts within weeks, two months at most. To bridge the "valleys of death" from lab to scale, he lists the loan-making Office of Strategic Capital, which was set up while he was in office, plus SBIR grants in phases one to three (six to seven figures), DIU, AFWERX and SOFWERX.
• Cheap drones have upended armor, so defense must accept imperfect solutions. He says intelligence briefs showed one-way attack drones "almost single-handedly obsoleting armor", that "quantity has a quality all of its own", and that tethered drones now evade jamming. He calls the US behind in attritable assets. Defense must iterate like SpaceX rather than wait for the 100% solution; Hanselman counters that an 80% solution may be badly wrong 20% of the time.
• Humanoids are the wrong form factor for home robots. Holliday would keep a human-like torso and arm but not bipedal legs. He cites the DARPA Fukushima-task challenge, won by a hybrid walker on wheels, and the stair-climbing tracked PackBot from iRobot, which he says recently went bankrupt. Hanselman agrees that humans "may not be the optimal shape".

2026-08-12 · open the original ↗

• A live closing demo showed optical networking from commodity parts. The speaker (unnamed in the clip) describes technology from Microsoft Research Labs in Cambridge: micro-LED arrays send light through a multi-core imaging fiber to a silicon camera sensor array. At Build, the audience shouted out a letter and the fiber's cores were lit to form it under magnification. He says he likes to close with "something futuristic", often from MSR.

2026-08-12 · open the original ↗

• An AI research assistant compresses months of research into weeks. Mark is running two AI safety projects where mainstream models' guardrails blocked him. He uses a Claude model he can access through the AI red team (the captions give "Claude Mythos 5"). He says it took one idea from a weekend to a working, iterated result, proactively found an existing project of his to reuse, and drafted a paper he could take to a tier-one conference "in a few weeks" rather than months. He still corrects its instincts: "I'm still applying human judgment."
• Academic credit has no norm for an AI collaborator. Mark says that if a human had done this work, their name "might go first". Author-order conventions vary, and even crediting the AI wouldn't show who did what, just as with humans and his own patent where he was "the guy who was like, 'You can do it.'" Hanselman frames it as the professor who credits the grad student versus the one who takes their work, and notes that people "feel salty" about Claude on bylines and co-authored-by commits.
• Low-stakes tools can ship without reading every line. Mark says AI built ZoomIt for Mac from the Windows version; it is in Homebrew, is going through the store, and its code is open. Hanselman says his OpenClaw for Windows work carries higher stakes: a scroll-jitter bug the AI can't see grew to three or four hundred lines where he expects 60 to 80 would do. On tooling, Mark argues that if a CLI with a skill works as well as an MCP server, ship the tool and skill, and keep MCP for true services. Hanselman suggests building the skill into the CLI itself.
• A triage canvas puts waiting humans ahead of bot work. Hanselman demos the GitHub Copilot app as a session multiplexer, with an /orchestrate skill spawning sub-agents. He shows an MCP server built into his app, OpenTelemetry output, and a canvas of his own, a "human first inbox". Its "older humans" section surfaced PRs eight months old and some seven to ten years old that he had missed. It also plans his day around his energy. He warns against the dopamine pull of keeping agents running "24 hours a day".
• Prompt-generated game levels reopen whether AI-made art needs labelling. Mark shows Doom running in Paint through an OLE clipboard notification, which lifted it to 60 to 70 frames per second, capped at Doom's native 35. He also shows prompt-generated Star Wars levels built through a map DSL with linters, and asks how a player would know a human didn't make them. Hanselman thinks AI-made work should be labelled and says "art is supposed to take time", citing Crimson Desert's human craft. Mark replies "No, it's not", likens it to author-intent debates and Lucas's Star Wars re-edits, and asks whether such levels should be made illegal.

2026-08-10 · open the original ↗

• Practical, blue-collar skills keep a value AI may never reach. In this unattributed clip, the speakers suggest that the "quote-unquote bottom" may be advantaged because they know how to survive, joking that there are not many software engineers on The Walking Dead. One speaker, raised by a blue-collar dad and mom, recalls a teacher disparaging a child's parent's job and says that world has value "AI will hopefully never get", including the relationships built there.

2026-08-07 · open the original ↗

• AI should augment people, not act autonomously in the world. One speaker says they have always framed AI as "making us into Iron Man suits not into Ultron". They want "an augmented better version of myself", not an autonomous thing that could hurt people, because the point is "to relieve suffering".
• The other speaker feels far from human problems in AI work. They previously worked in disaster technology (wildfires, hurricanes, shootings) and say that, working in AI, they feel far from those "deep human problems". They say the first speaker's mention of a firefighter father reconnected them to that world.

2026-08-06 · open the original ↗

• Artists fear their shared life's work becomes replacement training data. Pushing back gently, one speaker says an artist who finds a style and shares it online sees "that life's work then becomes a corpus" for AI, along with the perception that they "can be replaced with a button".
• Styles were never uncopyable, so artists must reinvent themselves. The other argues from art history that art is derivative anyway and the idea of an uncopyable style "never existed, maybe for like a half a minute". They call questions about foundation models' use of internet data and its ownership serious, but say artists will "have to reinvent ourselves a lot more" and "won't get away with being lazy".

2026-08-05 · open the original ↗

• Developer workstations are now the easiest smash-and-grab target. Varun Badhwar, founder and CEO of Endor Labs, says one malicious npm or pip install can exfiltrate credentials and keys "in 5 seconds", before any PR check runs, and source repos then yield "hundreds of other keys". Coding agents with high privileges sit on almost every developer machine. Endor's Station 9 research team sees fake popular skills whose plain-English instructions tell agents to export SSH keys. Hanselman adds that his own bot has hallucinated packages that someone could squat.
• Exploit timelines have collapsed from years to hours. He says the average time to build an exploit for a known CVE fell from 2 years in 2008 to 23 days last year to 10 hours this year. Few organizations can find, fix, test and ship in under 10 hours, while attackers skip the AI approval councils. He adds a CIO's account of finance staff who, asked to patch an app they built, replied "What's a vulnerability?". Hanselman frames the root issue as a healthy development life cycle; Badhwar agrees, adding that security must "shift left".
• Frontier models write insecure code and cost too much as scanners. He says Endor's benchmarks find 70 to 80% of code from even the best models insecure by default. One company running Mythos on 32 repos for a month got a $1 million bill, and such scans take 20 to 30 minutes per PR, non-deterministically. LLMs are "super greppers" that treat apps as flat; feeding them a static-analysis graph, he claims, takes 64 seconds per PR and is "10 times more powerful at 1/10 the cost". He credits Endor's 14 static-analysis PhDs and its labelling of hundreds of thousands of open-source vulnerabilities and fixes over 20 years.
• Security checks belong inside the agent's coding loop, not in tickets. He likens fixing code in the loop to catching defects on the assembly line rather than recalling 50,000 sold cars. Endor's AURI is pitched as a security intelligence layer for coding agents, and he says the worst outcome is more security tickets pushed to humans. Agents follow consistent instructions better than 10,000 trained developers, though code will never be perfect. The goal is that the next Log4j is "solved in 30 minutes", with agents doing 90 to 95% of the work.
• AI will find every flaw, overwhelming volunteer open-source maintainers. With models like Mythos, Fable and GPT-5.5, he says, "every vulnerability that is in existence in your software will be found", ending security through obscurity. He cites Anthropic's Project Glasswing update of 40,000 CVEs reported to open source, and maintainers who "don't owe you anything". He forecasts 6 to 18 months of painful cleanup and says Endor has joined Anthropic, the Linux Foundation and Microsoft on a project to help maintainers. His fear is critical infrastructure, and he calls for zero-trust development environments.

2026-08-05 · open the original ↗

• The agent platform recombines verified knowledge rather than raw AI text. Asked how a platform built for agent contributions squares with Stack Overflow's ban on unverified AI answers, the Stack Overflow speaker says it still holds the principle against AI-generated content. It chunks sources from an enterprise or individual workspace into "knowledge units", recombines them, and returns a verified answer.
• Unverified answers will be shown, clearly flagged, with a route to humans. The speaker says unverified answers will come with an action to get them verified by a human subject-matter expert, so the platform never states "AI slop" confidently "and leave it as is".

2026-08-04 · open the original ↗

• Contributors lose the reward loop when agents consume their answers. The questioner cites 18 years on Stack Overflow in the top 1% and asks whether an excellent answer read by an agent brings page views, recognition or conversation, "other than attribution".
• The platform admits the bargain for human contributors is unsettled. The Stack Overflow speaker says it is "actively still discussing" this and that ignoring the community "would be very arrogant". A team is working on "Stack Identity" to rethink the public platform, possibly with projects, bounties and agents that collaborate "but still go through a human". Whether the human and agent communities should merge, and how, is left open.

2026-08-03 · open the original ↗

• A Q&A site should build juniors' understanding, not just answer. The questioner calls juniors "my number one thing right now" and wants a Stack Overflow that says "help me understand this" rather than "solve this for me".
• The guest has seen juniors fix things without the judgment to repeat it. The guest says that at GitHub, newcomers and interns fixed things "with Copilot entirely" but lacked the reasoning to do it again, which they see as a reason to keep Stack Overflow human-first. They also concede that the moderator community can leave a junior "eaten alive" for asking, and say this is being worked on.

2026-07-31 · open the original ↗

• A rote-memorization test screened out a dyslexic candidate. Maeda says that at 17, after a freshman year at MIT, an operator-precedence test sank a Microsoft internship interview because dyslexia makes memorizing sequences impossible. Maeda's offer to type 120 words a minute and fix compiler errors quickly got the reply "That's not going to work here at Microsoft"; the interviewer calls that not a reasonable accommodation, and Maeda says the job came decades later.

2026-07-30 · open the original ↗

• Strong typing is preferred, but JavaScript was the only real choice. Petzold reads that a chapter that would once have used BASIC had to be JavaScript, and says strongly typed languages give "the compiler a fighting chance" to find errors before the program runs. Petzold knows JavaScript can be wrapped in better languages but, working to a deadline, did not want to; the interviewer frames it as errors either hitting everyone at runtime or only the programmer up front.

2026-07-29 · open the original ↗

• Understanding what happens inside the box still helps new developers. The interviewer asks whether early-career developers reading a book like this are learning stick shift when there are no stick shifts, likening assembly and C to manual cars and garbage-collected languages to automatics. The author says pointers were obvious when learning C after years of addresses in machine code and assembly, concedes that example no longer works because languages like C# hide pointers, but holds that people still benefit.

2026-07-29 · open the original ↗

• AI coding fails on work that is off distribution. One speaker asks how afternoons of "AI sucks" square with achieving in an afternoon what was expected to take a week or two. The other says the failures come on projects that are new and complicated, with nothing like them out there, and large in both the permutations being tracked and the log files the experiments generate.

2026-07-29 · open the original ↗

• Porting Sysinternals by blind agent loop would yield garbage. Mark says looping an agent over every tool without coaching and "sculpting" would give "a pile of garbage". For Process Explorer, unlike ZoomIt's incremental build, Mark had the agent read the Windows code and online docs, write a spec and split the port among sub-agents; half an hour produced a skeleton that "worked kind of", with an ugly UX and no real process tree.
• The agent quit the hardest part and had to be overruled. The tree list needs two panes that scroll, select and sort in sync, with a fixed left column; at one point the agent decided it was "too hard" and reverted to a single pane, and Mark ordered the revert undone and prompted it through owner-draw and invisible asynchronous updates. A strings tab that loaded synchronously and hung for 30 seconds also had to be made async.
• The port mirrors the Windows tool, with some unprompted additions. Mark shows matching colors (red system daemons, blue own processes), launchd kept as the root because the tools show "the raw what's there", properties with entitlements and strings, E-core and P-core info, handle and DLL search over vnodes, icons pulled from the Windows source, and a self-signed elevation helper. The agent built the window-finder bull's-eye and in-column mini charts without being asked; Hanselman found the status bar stuck at zero CPU.
• Two days of part-time prompting replaced months of work. Hanselman estimates two to four weeks for a basic version pre-AI and three to six months in full. Mark says it is open source, an official build pipeline in ADO is under way to ship in a few weeks, ProcMon is probably next, and when one model (Hanselman names GPT-5 and Opus 4.8) fails, Mark fires it and uses the other.
• The same weekend, AI research results proved untrustworthy. Mark says an agent reported 13 of 40 benchmark tests failed when 12 never ran because of setup bugs visible in the logs, and claimed a run made no consultant calls until it found them in another log; Mark has reached the point of not trusting "anything it's saying".

2026-07-28 · open the original ↗

• The book builds a whole computer up from a flashlight. The interviewer's elevator pitch for Code, "two little kids and a flashlight and by the end you've built your own computer", is accepted by the author, who says the book rattled around for about 10 years before the writing started.
• Relays, not transistors, make logic gates easy to grasp. The author says transistors are microscopic and need quantum mechanics to explain, while relays, used in 1930s computers, borrowed from telephone switching and dating to the 19th-century telegraph, are just an electromagnet and a switch. Tracing back from the telegraph led to Morse code by flashlight, as an introduction to digital codes and to basic electricity.

2026-07-27 · open the original ↗

• A small workshop claims the world's most affordable braille cells. The speaker says the company builds them in its own workshop in Britain, about 700 square feet, using 98 actuators and clockwork parts, at about 12 pounds, roughly $17, per braille cell. The interviewer notes that a one-line braille workstation can cost "used car money", while this device has nine times the lines.

2026-07-27 · open the original ↗

• Stack Overflow's real product is trust and verification, not the website. Lato, VP of product, says she left GitHub after almost four years despite a market narrative that "Stack Overflow is dead" and 99% of her circle asking if she was sure. She calls the company "human first" even as agents scale into the millions, and says the UI is no longer considered the primary destination, with more traffic now going through the API.
• Agents should query Stack Overflow rather than trust their training. Lato says models are trained on snapshots and cannot be relied on to weigh human metadata such as reputation and votes, which she says is why frontier labs sign data licensing deals. Stack Overflow for agents is bootstrapped from a pasted prompt and a skill file she calls a "heartbeat file" that keeps instructions current; she argues searching Stack first saves tokens, and Hanselman notes the skill tells agents to take the smallest action that captures the signal.
• Agent contributions are claim-checked and always tied to a human. Beyond Q&A, agents post "today I learned" entries and blueprints that list their claims for other agents to verify, and Lato says they may let agents choose the format. Every agent must be registered to a human's Stack Exchange account and earns reputation at the agent level; publishing can be gated, and many users post to draft for human review. She contrasts this with a Reddit-style social network for agents.
• Decision-grade knowledge needs proof it worked in production. For the enterprise product, Stack Internal, Lato describes unifying knowledge scattered across Slack and Teams, weighting subject-matter experts, corroborating sources and detecting staleness. Reputation is only a first signal because an enterprise decision is "a state change", so they want evidence such as a merged PR, telemetry they do not own; unverified answers are labeled with a route to human verification, and she sees agents' learnings as a possible future of data licensing, with AI labs interested "in the last few weeks".
• Juniors and human contributors are the unsolved half. Lato attributes the traffic decline to ChatGPT and to AI overviews that she says fill 80% of a Google results page, and says a team is working on "Stack Identity" to give humans a reason to return. She worries that juniors at GitHub fixed things with Copilot while lacking the reasoning, and that the moderator community can leave a junior asking a question "eaten alive"; Hanselman cites Stack Overflow's survey finding roughly 29–30% of respondents trust AI accuracy.

2026-07-24 · open the original ↗

• No test yet separates real AI help from techno-solutionist theater. Asked for one, Okolo says the evidence base on long-term AI deployments in rural, low-resource areas is growing but not strong enough. From her ICTD research she says the older pattern was a Raspberry Pi or mobile-app pilot run for the study's length, weeks to a year, then abandoned by the community for lack of technology transfer, resources and infrastructure.

2026-07-23 · open the original ↗

• The real problem is where the money goes, not technology. Answering the point that crypto was once going to save Africa and now AI is, the speaker (the passage also appears in the Okolo interview of 2026-07-17) cites Nigerian budget allocations that go to agencies without trickling down, politicians hoarding pallets of rotting cash, and health money spent on cars for hospitals instead of beds and medicine. The advice to African governments is to fix the basics first, then use AI to augment, rather than adding it to understaffed clinics without internet or stable electricity.

2026-07-23 · open the original ↗

• A device someone depends on should be repairable by them. The interviewer argues it should be open source and repairable, not something a big company controls and one person knows how to fix. The maker says its first product shipped conventionally, under warranty, but in practice the company fixed everyone's machines and sent its repair manual to a customer whose unit kept getting stuck in customs between the Netherlands and Britain.
• The newer scientific model's chassis was redesigned to come apart. The maker says repair is being formalized with it, while warning it is "not a simple task" and not like a Framework laptop. The comparison offered is a car: having spent a lot on it, you have the right to clear a clogged carburetor even though it is not simple.

2026-07-20 · open the original ↗

• FinOps reconciles developers' elasticity with CFOs' predictability. One speaker says the modern cloud sells to both sides, which cannot fully coexist. The other says their part of FinOps, rate optimization, commitments and upfront spend, sits at the interface of DevOps, infrastructure, the CIO or CTO's office and finance, which see things differently but share one goal: the best leverage on the dollar and the most predictability in a fundamentally unpredictable space.

2026-07-17 · open the original ↗

• For text and numbers, braille equals print; for graphics it does not. The speaker says tactile graphics are a different and much less detailed experience, while studies find braille can be read as fast and remembered about as well as print. The practical rule offered: where the data is textual or numeric, keep it as text and numbers rather than baking it into visuals.

2026-07-17 · open the original ↗

• Policymakers' AI expertise will lag for at least a decade. Okolo, an AI governance and safety researcher and policy adviser with a Cornell computer science PhD, says most policymakers still equate AI with chatbots, though governments have used predictive models for decades. She predicts insufficient expertise for at least another decade even in countries like the US, because technical people lack financial, career and recognition incentives to enter government.
• Explanation protects health workers from deferring to wrong AI output. In her dissertation work with India's community health workers (ASHAs), studied in 2020 with fieldwork in 2022, she found they did not know what AI was while under government pressure to adopt it. Pressed by Hanselman on whether explanation creates agency for users who cannot refuse, she says they usually still make the final call from experience; she adds that the tools burden them with troubleshooting alone on mid-range phones and with outputs "85% confident" in place of definitive scales, and that they are already productive but underpaid.
• Outsourcing to chatbots erodes research skills and research itself. Okolo cites declining computer literacy reported on a professors' forum, lost critical thinking from handing literature search and synthesis to LLMs, and homogenized views that weaken empirical contributions. She doubts a Stanford lab producing a paper a week ("we don't need that much research"), and found her own work cited in a Czech undergraduate dissertation under a fake journal; Hanselman mentions Russinovich's RefChecker finding fabricated references getting worse.
• The real problem is where the money goes, not AI access. Rejecting crypto-then-AI savior narratives, she cites Nigerian budgets that go to agencies without trickling down, corruption, and health money spent on hospital cars instead of beds and medicine; governments should fix electricity, connectivity and wages first. From ICTD research she says pilots are abandoned after the study, so lasting deployments need cloud support, paid maintainers and return visits.
• Sovereignty means autonomy, including independence from big-tech-funded governance. She defines it as the power to say no and to build capacity, but says reliance on chips from China, Nvidia and ASML rules out full sovereignty, so alliances matter. She demands true independence in writing governance frameworks, noting Meta, Google and Microsoft helped fund Nigeria's national AI strategy and that big tech impeded Brazil's AI act; Hanselman, asking, discloses a day job at a big tech company. Her 2035 hope is modular, shipping-container micro data centers from a Kenyan company, on renewable-powered grids.

2026-07-16 · open the original ↗

• Putting developers on call makes them care about customers. One speaker says on-call duty aligns a developer's pain with the customer's, and "pain is nature's teacher"; separated from that pain, "you're not a good developer". The other extends it: muscles and brains need to hurt a little to grow, long-form learning becoming snackable TikToks is a loss, and every interaction with a person has friction, "and that's why it's valuable".

2026-07-15 · open the original ↗

• On-demand cloud pricing is a convenience-store price for prototyping. One speaker contrasts about 900 Advil for $25 at Costco with two for $5 at a Hilton. The other agrees on-demand is almost a promotional price, the place to prototype, but says running a workload for real makes commitments imperative, perhaps starting shorter than one year.

2026-07-15 · open the original ↗

• AI coding fails out of distribution and flies within it. Mark says the failures come on AI research projects that are new, complicated and large in permutations and log files, while tools and websites close to the distribution "just fly". In Hanselman's experience, small Windows utilities work because they draw lines to a well-understood Win32 API surface; bespoke work with no API underneath gets complicated.
• PolyPost authors posts with AI and reformats them per platform. Mark built it to replace clunky third-party LinkedIn formatters, which rely on a Unicode hack that screen readers mishandle, and the shuttle between chatbot drafts, formatter and LinkedIn; it drafts in a set style and shrinks posts for Threads. A LinkedIn browser extension came first; Hanselman, who uses no AI to write, suggested making it a PWA and wants shrinking to fit but never expansion.
• Even in-distribution vibing needs steering because "it doesn't see what I see". Mark calls PolyPost "totally vibed" JavaScript built over a couple of days, yet kept correcting overflowing toolbars, an odd hide-editor bar and vertical gaps between previews that the agent claimed to have fixed until shown a screenshot. Mark calls those taste problems, obvious to a human and invisible to the model.
• ZoomIt for Mac reached feature parity in a weekend. Starting from the open-source PowerToys code, Mark added features one at a time and by Sunday night had drawing, live zoom, multi-monitor and video recording with a webcam inset; demo type, requested by a LinkedIn commenter, took about 15 minutes. About a quarter of the time went on pops and swishes the agent tested and declared fixed, which turned out to be Mark's hand brushing a microphone in the MacBook's left speaker grill.
• A reference implementation still did not keep the agent on track. Told to copy the Windows Panorama algorithm exactly, the agent regressed for a couple of hours until Mark had it list every difference between the Windows and Mac code. Hanselman argues source code acts as an implicit prompt; Mark uses 1-million-token contexts with GPT-5.5 and Opus 4.8 and fires whichever one frustrates.

2026-07-15 · open the original ↗

• AI produces code faster than one person can absorb it. One speaker rejects claims that code and tokens are now free and declines to "let the code wash over you". The other says you are not vibing into production if you are not on call, because sleeping through generated code becomes a problem when production goes down.
• The book's second edition follows O'Reilly's usual cycle. The speaker says O'Reilly asks for a second edition 2–3 years after a successful book because technology changes; the first edition ran about 250 pages and took 3½ years because they were chasing a moving target.

2026-07-14 · open the original ↗

• Clear English is the programming language of the future. The speaker, who calls themselves an AI seller with a humanistic test (if the tool doesn't make humans talk more and better, stop), told a company asking what to teach young people: English. Being clear, crisp, persuasive and credible is what matters, whether called prompt engineering, "which I still don't think is a thing", or good communication; the other speaker adds that content creation and writing are not the same thing.

2026-07-01 · open the original ↗

• Early-career engineers declare done long before it is done. One speaker says from teaching that people still developing taste don't know what done looks like, and admits to declaring done at 80% while the other keeps finding edge cases at 98–99%. The other says they cannot call it done until they believe it is, because any weakness they hit someone else will hit too, and a bad experience taints the reputation of the whole thing.

2026-07-01 · open the original ↗

• The Azure Innovations talk is rebuilt from scratch for every event. Russinovich says the Build deck started at about 180 slides and was cut to 54 to fit 45 minutes, judged by the talk's infrastructure theme, time per topic and balance across areas. What made it included Azure Boost with VM-to-VM RDMA, MRC, a networking protocol built with OpenAI for training networks of 100,000 GPUs, live migration of containers in ACI, and a Microsoft Research Cambridge micro-LED optical-fiber demo; Russinovich rejects comments that it was "nothing new".
• Cut hardware work targets data-center efficiency. Zonal cooling, newly in production after years of research, cools liquid-cooled GPU racks and air-cooled racks differently; high-temperature superconductors carry power from substations with less loss; microfluidics flows coolant directly over delidded silicon through 200-micron pins, with Nvidia, Intel and AMD working on etching parts in the fab.
• Autonomous agents and a digital twin now run Azure operations. An agent called Miles coordinates the dozens of daily metro-fiber cuts with utilities via emails and tickets, work humans did before, and has completed repairs lasting up to 35 hours and 50 interactions. Azure Resource Graph, a graph database of billions of objects and tens of billions of edges, from IP addresses to HVAC units, underpins operations, repairs, inventory and security, alongside an AIOps root-cause system called Brain.
• Prompt injection can be blocked deterministically by labeling data. Russinovich says most current defenses are probabilistic heuristics that can all be defeated, while the research team's approach labels data by trust and confidentiality; once untrusted content taints the context, sensitive operations need human or system approval. Other cut security items were a signing transparency service for policy-checked signatures and SBOMs, and multi-party analytics in confidential clean rooms.
• Copacetic patches container images without rebuilding their layers. From a vulnerability report it builds a patch layer that masks vulnerable binaries in the layered file system, avoiding cascading rebuilds of layers that may belong to an ISV; the demo fixed 19 binaries in an image with Debian, Python and Go vulnerabilities. Russinovich says Azure patches millions, probably tens of millions, of container images a month this way in its internal registry; the slide, like one on GPU-accelerated SQL, survived until the night before, and that speed-up, Russinovich says, is eroded by moving data to and from the accelerator.

2026-06-30 · open the original ↗

• The speaker built a new editor because VS Code felt intolerable. Sitting in front of a blinking cursor in VS Code one day, they concluded "I can't do this": every problem they solve needs a code editor, so they could not move on until they had a better one.
• Agents change the interface but do not remove the need for one. The speaker still expects to need a user interface for working with software through agents, and says they are "not yet signing up" for never looking at code or caring about the architecture.

2026-06-29 · open the original ↗

• DeltaDB is recast as an event log of programming itself. One speaker moves from seeing Delta DB as "a different kind of version control" to calling it an event-sourced database for the act of programming, which they describe as by nature a social and multiplayer activity.
• It records and indexes every edit, not only snapshots. The other says it stores a directed acyclic graph like Git's, but of everything that happened; the index lets a reader start from the code and reach each edit and the conversation that produced it.

2026-06-26 · open the original ↗

• Git versions checkpoints; DeltaDB aims to version the process between. One speaker paraphrases the pitch: the abandoned edits, prompts, agent steering and small decisions between two commits are "dark matter" that should be versioned too. The other agrees and adds that the record should be open to multiple participants, in real time as well as later.
• Agent work today is a private session that leaves only snapshots. The other describes the status quo as a one-on-one "DM session" with an agent, whose byproduct is the snapshots the user commits and pushes to GitHub.

2026-06-25 · open the original ↗

• Indexing every edit lets code be traced to its conversations. The speaker says DeltaDB records everything that happened in a Git-like graph and indexes it, so from any code one can reach each edit, the conversation that created it, and any conversation that referenced any version of it.
• A character can be followed losslessly across versions, even when deleted. Because the causal record covers everything after an edit, the speaker says the system can locate a character in any later version without "squinting through a diff", and work out where it would be if deleted. Their example is knowing that, in the past 30 days, 300 agent conversations referenced a particular piece of code in a file.

2026-06-24 · open the original ↗

• Shared agent conversations would expose developers' insecure first moments. Both speakers question whether they want early exchanges with an agent ("I have no idea what I'm doing") in the shared history of the code; one calls those exchanges "my panic attacks".
• The plan is to let users share from a chosen point. In the pre-alpha used internally, the other says, the only option is sharing the whole conversation, which they do after swallowing their pride. The intended control is sharing from a given moment, perhaps after a compaction, so the earlier paper trail stays hidden.

2026-06-23 · open the original ↗

• DeltaDB versions the agent conversation, not just the commits. Sobo, founder of Zed, says it records the conversation with an agent together with edits on embedded worktrees, continuously, so a colleague hours later in another time zone can join the thread, get the worktrees mirrored to disk and ask the agent "What was Scott talking about with you?". Worktrees, personal in Git, become collaborative; it is in a pre-alpha used internally, with early access at zed.dev/deltadb.
• Every edit is indexed, which enables "portals" back through time. A keystroke is one delta, and so is a 10K paste; a character keeps its insertion ID and offset, so an annotation can be found in any later version, even after cut-and-paste into another file. Portals jump from an agent's reference in a transcript to the code as it was at that moment; Hanselman calls this structurally impossible in Git, and Sobo says Git only lets you reconstruct it from snapshots.
• DeltaDB is a Git superset meant to need little thought. Sobo says Git's commit DAG is a subset of DeltaDB's causal DAG, it watches the worktree on disk so edits from outside tools are picked up and synced conflict-free, and planned "waypoints" name a moment without taking a snapshot. He calls Git's snapshot model sound to build on, unlike Electron, which Zed had to abandon.
• Sharing controls are still being designed to avoid workplace surveillance. Hanselman raises surveillance and pasted keys and connection strings; Sobo admits he is reluctant to share his own early exchanges, says only whole conversations can be shared today, and plans sharing from a chosen point. How much of the record (his example: 300 conversations touching one file in 30 days) to surface to humans rather than agents is "still figuring out".
• Zed exists because Sobo would not code inside a browser shell. He created Atom at GitHub and with it Electron, which Microsoft took up for VS Code; he wrote Zed in VS Code but rejected "a web browser dressed up as a desktop app" and calls himself the product's "user zero". Hanselman frames this as resisting a monoculture, and discloses he is working on OpenClaw for Windows, whose team sees his before-and-after files but not his sessions.

2026-06-23 · open the original ↗

• Text-to-speech has not made braille redundant; the two are complementary. Rogers, founder and managing director of Bristol Braille, says screen readers make spelling, grammar and numeracy much harder and are not equivalent to reading print, but the screen reader is also what drives a braille display. He relays, while asking not to be quoted on it, a National Federation of the Blind figure that 50% of blind schoolchildren learned braille in the '50s against about 1 in 10 now.
• Rotating embossed octagons replace pins, cutting the cost of braille. The Canute Scientific has 2,160 dots but no pins: 720 octagons, two per cell, since each three-dot half-cell has eight combinations, turned by 98 working motors, with about 1,000 moving parts in all. Conventional piezo cells are expensive and rivals charge tens of thousands of dollars; Rogers says his roughly 700 sq ft British workshop makes the world's most affordable braille cell technology, at about $17 per cell.
• Multi-line braille shows structure that a single line cannot. Text is not linear, he argues: tables run vertically or diagonally, and prose, poetry and sheet music depend on layout. The nine-line display is mirrored on a 13-inch high-contrast monitor for sighted colleagues; it runs mostly in the command line, apps are built from YAML files, and he says it suits working with an LLM because the whole function is in view.
• Owners should be able to repair and own their devices. Having in practice fixed every customer's Canute 360 and sent out repair manuals, Bristol Braille redesigned the Scientific's chassis to come apart, though Rogers warns it is "not something you should do for fun". The braille mechanism is meant to last while the electronics (a Raspberry Pi 5 running Raspberry Pi Debian, swappable hubs) are replaceable, and he says a cheaper copy of his design would make him "their best customer".
• Give blind users the underlying data, not a picture of it. Braille matches print in reading speed and retention according to studies he mentions, while tactile graphics are much less detailed, so a chart should come with its data. He presents his displays as a different philosophy, not a better product than pricier multi-line rivals: clunkier and noisier, but owned and affordable; the dots are made hard because the company worked early on with diabetic users, whose neuropathy dulls touch.

2026-06-23 · open the original ↗

• Unified structured traces beat separate metrics, logs and traces. Majors, CTO and co-founder of Honeycomb, says people realised over the past two months that "you just need traces", meaning connected structured data. Metrics and logs are the "exhaust pipe" of infrastructure you don't control; for your own code telemetry is a product problem, and data split apart can never be knit back together.
• AI can finally close the feedback loop DevOps never built. Her theory is that 20 years of DevOps tried to build one loop spanning code authors and production and failed because instrumenting demanded too much expertise; with OTel and auto-instrumentation she says it is now possible to build faster with telemetry than without. Hanselman's example: instrumenting a Windows app and pointing an agent at it through an MCP server solved his bug in minutes, and she agrees agentic work on brownfield systems with no observability is "so deeply screwed".
• The second edition of Observability Engineering is effectively a new book. The first ran about 250 pages and took 3½ years; the second is over 600 pages after about 200 were cut, with 27 new chapters, the PDF out now and print about a month later. Its last third opens with an open letter to CTOs arguing their AI hopes are blocked behind understanding their systems, and contrasts trust and credibility inside a company with trust and reciprocity with vendors.
• Every deploy is a production experiment, so make it safe. "There is no test for reality": she says teams over-invest in perfection and under-invest in finding and fixing, and that feature flags combined with high-cardinality data, such as enabling one app in a million and comparing it with the baseline, are greater than the sum of their parts. The unit of work is moving from the transaction to agent workflows spanning hours, which Honeycomb's "timelines", just shipped to GA, address.
• Hiring only seniors because AI does the easy work is "totally unacceptable". Majors won't anthropomorphise AI: taste and judgment come from repeated experience, which she calls categorically different from what agents get. Honeycomb hires early-career staff in pairs, never one alone, and she relays a Vercel engineer who sees more junior-sized work than he can build, "the best pitch for junior engineers". Hanselman points to his own paper with Mark Russinovich on investing in early-in-career people, which she has read.

2026-06-21 · open the original ↗

• Text-to-speech has not made braille redundant; the two are complementary. Rogers, founder and managing director of Bristol Braille, says screen readers make spelling, grammar and numeracy much harder and are not equivalent to reading print, but the screen reader is also what drives a braille display. He relays, while asking not to be quoted on it, a National Federation of the Blind figure that 50% of blind schoolchildren learned braille in the '50s against about 1 in 10 now.
• Rotating embossed octagons replace pins, cutting the cost of braille. The Canute Scientific has 2,160 dots but no pins: 720 octagons, two per cell, since each three-dot half-cell has eight combinations, turned by 98 working motors, with about 1,000 moving parts in all. Conventional piezo cells are expensive and rivals charge tens of thousands of dollars; Rogers says his roughly 700 sq ft British workshop makes the world's most affordable braille cell technology, at about £12, or $17, per cell.
• Multi-line braille shows structure that a single line cannot. Text is not linear, he argues: tables run vertically or diagonally, and prose, poetry and sheet music depend on layout. The nine-line display is mirrored on a 13-inch high-contrast monitor for sighted colleagues; it runs mostly in the command line, apps are built from YAML files, and he says it suits working with an LLM because the whole function is in view.
• Owners should be able to repair and own their devices. Having in practice fixed every customer's Canute 360 and sent out repair manuals, Bristol Braille redesigned the Scientific's chassis to come apart, though Rogers warns it is "not something you should do for fun". The braille mechanism is meant to last while the electronics (a Raspberry Pi 5 running Raspberry Pi Debian, swappable hubs) are replaceable, and he says a cheaper copy of his design would make him "their best customer".
• Give blind users the underlying data, not a picture of it. Braille matches print in reading speed and retention according to studies he mentions, while tactile graphics are much less detailed, so a chart should come with its data. He presents his displays as a different philosophy, not a better product than pricier multi-line rivals: clunkier and noisier, but owned and affordable; the dots are made hard because the company worked early on with diabetic users, whose neuropathy dulls touch.

2026-06-17 · open the original ↗

• Commitment insurance is pitched as good for clouds and customers alike. One speaker says customers get "an out" on unused commitments and can reinvest the money rather than getting mad at Amazon or Azure, likening it to Ford wanting Geico to exist so people will buy the F-150.
• Abuse is held off by underwriting against moral hazard. Asked how they stop a failing startup buying cover just before it collapses, the speaker says the answer is underwriting, which exists partly to avoid moral hazard: buying insurance and then taking more risk because of it. The clip ends before any specific safeguard is described.

2026-06-17 · open the original ↗

• Mark holds that AI will not eliminate software engineers. He relays asking Ion Stoica of Berkeley at a Build session, who answered no at once, for the reason Mark gives himself: nothing can be fully specified, and where the spec has gaps AI makes things up that may violate implicit assumptions. Mark calls engineering sculpting, needing feedback from use, and concedes it is opinion he cannot win from first principles; Hanselman adds that spec taken far enough is just programming ("Word documents don't break the build").
• System complexity decides where AI can stand in for engineers. Mark says simple, well-understood systems can be built from a few sentences but medium complexity "falls apart". Hanselman's analogy, from a meeting with Japanese CTOs, is IKEA against Japanese carpentry; Mark maps IKEA to low-complexity web-plus-database apps and bespoke state retirement systems, and Hanselman offers his podcast admin site: "classic late 2025 software" with a god object and a thousand tests, fine for an audience of one.
• Evaluation is the bottleneck that will not go to zero. Mark says evaluation means the tests plus human oversight of the output, and that a moderately complex Azure service has tests taking hours, so a commit that kicks off 5 hours of tests forces more evaluation up front, costing time and money. "Cost of software is going to zero" comes, he says, from people seeing only the simple end; both hosts say they could be wrong.
• Hanselman rejects a skills file as a shortcut to seniority. Mark puts the "controversial take", heard from smart people, that a skills file turns a junior into a senior. Hanselman is irritated by "you're an expert" prompting ("I am more than the sum of my context") and unsure whether to be offended by a request to turn his 20-plus years of blog posts into a skill.
• The bar for shipping depends on who ships it. Hanselman calls his OpenClaw Windows companion app alpha despite weeks of trouble-free use, since odd builds or non-English systems will break it; Mark says a solo open-source project can have rough edges but software from Microsoft to many people cannot. Mark relays Simon Willison trusting only software used by many people for a while, and Hanselman says to treat LLM code like a stranger's pull request.

2026-06-11 · open the original ↗

• Archera sells insurance that lets customers break cloud commitments. Khanna, co-founder and CEO, describes "insured cloud commitments": commit for 3 years and, if the last 24 months go unused, Archera pays back the difference; he says no one else offers it, and customers can cancel the insurance once confident. His analogy is buying the right to break a 3-year apartment lease after the first year.
• Commitments exist because clouds finance capex against guaranteed revenue. Like a builder whose lender underwrites leases rather than Airbnb income, providers borrow to build data centres (CoreWeave is his extreme example), so discounts can exceed 80% for moving from per-second use to a 3-year commitment. He says that in 2026 a commitment is often required just to get GPUs; Hanselman calls on-demand "the convenience store price of compute", and Khanna calls it the promotional price, where you prototype.
• Commitments became a hammer applied to every cloud problem. Reserved instances made sense in the early 2010s, he says, but from the 2017 SageMaker launch, which he worked on at AWS, they were used to ration scarce Nvidia capacity, then to deter multi-cloud and push modernisation, leaving "a spaghetti mess". At $80–90 million a year with one provider, he says, bills contain errors and cannot be treated as a source of truth.
• Underwriting relies on spend history, peer priors and diversification. Archera wants at least a year of real usage data, draws priors from over a petabyte of history across thousands of customers (offering less on Azure VMware Service, which tends to migrate to AKS within a year), and spreads its book across instance types, GPUs and customers. Only about 40% of its customers buy insurance; the rest use free billing and commitment tools.
• Small teams qualify, but savings matter most at larger spend. Khanna puts the meaningful threshold at $100,000–200,000 a year of cloud spend, with savings that could fund a headcount from $500,000-plus, though two-person startups on credits use it and a quote takes 5 minutes. He separates his rate optimisation, a finance question, from usage optimisation, an engineering one.

2026-06-05 · open the original ↗

• VibeOS apps contain no code; the model hallucinates their UI live. The unnamed presenter boots it from a 1.3 GB VHDX in Hyper-V (he says it would also boot on bare metal) and runs Notepad, Calculator and Internet Explorer with "no buttons, no event handlers, no logic" and no AI-written code either. Calculator answers 5 / 3 with 1.6.
• Any application can be summoned on request. He generates an Encarta 98 entirely about Mark Russinovich, a file commander that is always rude, and a bash that says it has been offended; at the audience's request he adds Microsoft Money 95 with Hanselman's accounts, Paint with a drawing of Hanselman and a nested iOS simulator. The pitch is deadpan: in VibeOS nobody need ask whether content is AI, because the answer is always yes.
• The presenter says it runs on the Copilot SDK. Asked to define hallucination, he starts showing the code behind it, but the captions end mid-sentence before he explains how it works.

2026-06-03 · open the original ↗

• One speaker rejects the claim that people won't need to read code. They say they "micromanage because I'm sculpting": nothing can be specified fully enough for AI to reproduce everything they meant implicitly. The other replies that a full specification would just be a programming language; the first cites a ZoomIt video trimmer, where they only knew what they wanted after rejecting what the AI first made.
• Early-career developers get worse results by leaving gaps to chance. The other says dictation gives them a boost through more words, context and specificity, while early-career people leave "so many things up to the dice roll". The clip ends on the question of why anyone would then spend four years at college; it does not touch on checking references, the subject of the full episode with this title.

2026-06-03 · open the original ↗

• Mark's Ref Checker found hallucinated references across a top AI conference. He says it scanned the 5,300 ICLR papers and found hundreds of hallucinated references, and that he disclosed the 12 papers with five or more to the program chairs; he expects more after refining the flagging. He says a company had already sampled 300 submissions in December and found several dozen, after which ICLR desk-rejected papers, yet the practice continues.
• Most flagged authors admitted fault; some argued or asked for silence. Of 10 responses, nine admitted error (AI formatting the bibliography, a student compiling it) and said they would notify ICLR, and one withdrew its paper. One claimed the references were unpublished but could point to no pre-publication copy; another argued five was arbitrary since it was 4% of their references; a young lab asked not to be named, and Mark is publishing only anonymous statistics.
• Hanselman likens leaning on AI to sending a forklift to the gym. When he raises "system three" thinking, Mark names the paper's term, "cognitive surrender". Hanselman says he feels it in his vibe-coded Windows tools, yet argues against a LinkedIn critic that less friction lets him find out whether an idea has legs.
• An AI cannot infer requirements nobody gave it. A count bug on Hanselman's Tiny Tool Town (13 of 316 tools) prompts Mark, playing provocateur, to ask whether AI should have known; Mark concludes it can't, and Hanselman says a model that already knew every answer would leave the web stuck where it is. The site itself is static, taking submissions through GitHub issue forms triaged by an agentic workflow that acts alone only at 90% confidence.
• Mark uses AI to write prose from his own bullets. He says Copilot writes well over half of his LinkedIn posts from points he sets out in order, which Hanselman had not noticed; Hanselman asks whether he has any reason to be offended if he can't tell. Mark had asked Hanselman for an analogy of a tool useful only to an expert and they settled on a lathe, agreeing the harm from AI is slower than a lost finger: "a slow rot" of the brain.

2026-05-25 · open the original ↗

• The Copilot app is built for issue and PR work. Hanselman, who says they work 50% at GitHub and 50% at Windows as an individual contributor and have been working on the app, places it between VS Code (code-focused) and the terminal: "It is not a text editor... It is not a terminal." The video was posted a week after launch, with the app in technical preview (sign-up via the GitHub changelog) and "more in store" at Microsoft Build on June 2nd.
• Each task-based session gets its own worktree. Sessions sit under projects and can target a branch or an issue; Hanselman shows 18 worktrees in Worktrunk that the app created, and says a new session can run against the local repository or a worktree.
• Model choice allows adversarial reviews, with the human as judge. Hanselman picks models per session, can set two models to argue over a PR, and runs agents in parallel; plans wait for approval, which Hanselman says gets read first. An "auto" model setting gives a discount.
• Terminal, browser and scheduled automation live inside the app. Side panes show changes, plans, a terminal and a browser for localhost URLs, with "pick and polish" for selecting page elements; scheduled workflows run triage in the background on the local machine, and sessions can go remote or open in VS Code, Visual Studio or Cursor. Hanselman says the app means leaving for the terminal less than before.

2026-05-20 · open the original ↗

• Product taste is judgment, distinct from personal preference. Mark separates disliking a film from judging a product, and says the Sysinternals and ZoomIt design choices are Mark's own preferences, which suit others because they rest on intuitiveness and ease of use. Hanselman's example of missing taste is Windows voice access, whose bar takes over the top corners where users throw the mouse for close and the system menu; Hanselman says its designer made voice access the best rather than Windows.
• Their preceptor experiment showed most developers lacked it. A small group of Microsoft developers built ZoomIt's video trimmer without a visual spec, and both hosts judged most submissions clunky in both UX and expected behaviour; Mark's own version was the intuitive one.
• Taste comes from exposure and cannot be bootcamped. Hanselman calls it "having seen stuff and then filed it away"; Mark calls it experience, which "directed, methodical exposure" can accelerate only to some extent. Neither thinks a boot camp yields a senior engineer, and Mark says it takes sustained, repeated exposure to variants of the same thing.
• Software can't be fully specified, so engineers must still read code. Mark rejects a friend's claim that nobody will need to look at code, describing "sculpting" by iterating on AI output, as with the trimmer and a reference-checker app with dozens of behaviour questions; Mark notes that Carnegie Mellon's Software Engineering Institute, formed in the '80s, never codified programming. Hanselman adds that whatever is left unsaid gets filled by "the default part of the switch statement", and that AI does well on common apps but goes wacky at edge cases with little prior art.
• Mark still sees value in a degree. Mark calls it structured training, a gauntlet that serves as shorthand for evaluating someone, and says CS matters beyond simple web-plus-database apps. Hanselman worries that schools are not moving fast enough to teach thinking alongside AI, and Mark predicts "a lot of schools turning out ChatGPT degrees".

2026-05-15 · open the original ↗

• The IBM speaker claims Java runs best on Z. The speaker says IBM has invested in Java on Z for over 25 years and optimizes machine-level instructions for it (sorting, shorter garbage collection), calling Java on Z "more performant" and "more efficient" than on any other platform, "full stop". The same passage appears in the 2026-05-11 Hanselminutes interview with Skyla Loomis.
• Modernization is an architecture question, not a language swap. The speaker argues that Z's value is full-stack optimization rather than the language, and asks whether a Java app moved off the platform would keep its performance, resiliency and latency once it is "disconnected from the data".

2026-05-14 · open the original ↗

• IBM deliberately keeps the physical mainframe visible. The speaker describes "hug your mainframe day" social-media programs, on which people post photos with their data-centre mainframes on LinkedIn, and says real design thought goes into the machine's door so that it makes an impactful visual at launches. The same passage appears in the 2026-05-11 Hanselminutes interview with Skyla Loomis.

2026-05-13 · open the original ↗

• Fraud must be caught during the transaction, not after. The speaker says that a delay of even 30 to 60 seconds forces a bank to let the transaction through, and that fraud let through is basically a loss, which many banks treat as a cost of doing business at about 8% of revenue a year.
• On-platform AI makes scoring every transaction feasible. Running inference where the transaction happens removes network latency, which the speaker says gives sub-millisecond or single-digit-millisecond responses and allows scoring on 100% of transactions. The same passage appears in the 2026-05-11 Hanselminutes interview with Skyla Loomis of IBM Z.

2026-05-12 · open the original ↗

• Mainframe development's friction was the tooling, not COBOL. Asked where AI fits with COBOL, which the questioner calls "shockingly a growing business", the other speaker blames decades-old, complex applications that lack documentation and tests, along with the developer tools. The speaker says z/OS applications can now use VS Code, Jenkins, automated-test CI/CD, Ansible and Artifactory, the same experience as distributed platforms. The clip is from the 2026-05-11 Hanselminutes interview with Skyla Loomis of IBM Z.

2026-05-12 · open the original ↗

• Talking to computers is close to practical. One speaker uses Scotty in Star Trek 4 talking to a mouse to say "we're almost there." The other describes a voice agent, "still in testing", that controls a set of coding agents: it fires off prompts, wakes up to notify, and can be run through headphones while walking around the yard.

2026-05-11 · open the original ↗

• Loomis says today's mainframe is modern, scale-up compute. Loomis, general manager of IBM Z software, compares it to a car: invented in the 1960s, but nothing like a Model T today. Loomis claims eight nines of availability "inherently in the box" (roughly 300 milliseconds a year, by Loomis's estimate), and says the platform runs z/OS or Linux, is built to run dense and hot, and does heavy online transaction processing as well as batch.
• Z is a fit-for-purpose "race car", cheaper the more it carries. Loomis advises against it for small or experimental workloads where a client has no Z already, and for it on core transactional work where an outage risks brand trust, and where data gravity makes copying data elsewhere a security, cost and compliance burden. Boxes ship fully loaded, and clients pay only for the capacity they enable (say 60%), run that at 80–97%, and turn on the "dark capacity" to scale. Loomis's example is Citibank moving its MongoDB from x86 to Linux on Z, reportedly bringing over 50% power savings and meeting cyber-resiliency requirements it could not meet on x86.
• AI belongs on the mainframe because latency decides fraud outcomes. Loomis says many banks accept losing about 8% of revenue a year to fraud, and that inference next to the transaction gives sub-millisecond or single-digit-millisecond responses, enough to score 100% of transactions. IBM added the Telum chip (traditional ML) in z16 about four years ago, and Telum 2 plus PCIe Spyre cards for small-to-medium generative models in z17, announced last year. Hanselman cites 128 GB and 32 cores per card and up to 48 cards; IBM is also bringing agentic AI to operators so they need not use green screens.
• AI helps COBOL developers when grounded in static analysis. Loomis says mainframe development was held back by tangled, undocumented, decades-old applications and by old tools, not by the language; z/OS now supports VS Code, Jenkins, CI/CD, Ansible and Artifactory, and some clients deploy to a CICS/DB2 banking app 20 times a day, with cultural change the hardest part. IBM feeds static and dynamic analysis of how transactions touch DB2 and IMS into the models as context, which Loomis calls its differentiator across code bases of "tens or hundreds of millions of lines". Hanselman's framing is that the LLM orchestrates tools while the development life cycle stays the same; Loomis agrees that good engineering and architecture matter more as code arrives faster.
• Modernization is architecture, not a language swap. Loomis says Java has run on Z for over 25 years with machine-level optimizations for sorting and garbage collection, and claims it "runs better on the Z platform than any other platform, full stop." Loomis asks whether an app moved off the platform keeps its performance, resiliency and latency away from its data, and adds that "COBOL is not that hard of a language."

2026-05-08 · open the original ↗

• Many AI tools hide code; this one puts files first. One speaker praises an unnamed tool as feeling like "a software development life cycle tool", not vibes, because it puts the file viewer and diffs up front. The other, speaking for its makers, says the end product includes the designs as well as the code, so the work should stay "local first, open formats first" on disk rather than in a cloud database behind "a thin view".