{"slug": "claude", "name": "Claude", "category": "ai-coding", "category_name": "AI coding tools", "description": "Anthropic's AI assistant and family of language models, used for chat and agentic coding.", "recommenders": 648, "commenters": 1494, "warners": 217, "score": 529.59, "rising": 97.6, "first": "2025-11-01", "latest": "2026-10-31", "why": "Liked for writing and refactoring code, reverse engineering old software and hardware, and diagnosing problems from logs.", "why_is_quote": false, "has_verdict": true, "sentiment": {"recommend": 694, "neutral": 684, "compare": 76, "warn": 236}, "spark": [16, 25, 33, 39, 65, 36, 49, 93, 90, 91, 71, 86], "listed": true, "loved": ["Code quality", "Reverse engineering"], "caveat": "Security refusals", "short_quote": "claude basically is able to think critically while being neither sycophantic or argumentative most of the time", "icon": true, "domain": "claude.ai", "github_repo": null, "official_url": "https://claude.ai", "pricing": "freemium", "aliases": ["Claude", "Claude 5", "Claude 4.6", "Claude 4.5", "Claude 4.7", "Claude 4.8"], "verdict": [{"text": "Liked for writing and refactoring code,", "cites": [{"id": 804166944, "item_id": 50260434, "url": "https://news.ycombinator.com/item?id=50260434", "by": "misja111", "date": "2026-10-27", "stance": "recommend", "quote": "Claude has gotten so good that I have to admit it's a better programmer than myself.", "excerpt": "Until half a year ago I used to agree with you, but nowadays Claude has gotten so good that I have to admit it's a better programmer than myself. Its coding style is excellent, it follows whatever conventions are used in the codebase, it has excellent suggestions for refactorings and it spots omissions or mistakes in the specifications I give it. (I'm using mostly Scala as a language in a large microservices environm", "context": "works in a large microservices environment", "use_case": "programming in Scala microservices", "switched_from": "", "story": "The Mythical Man-Month in the Age of AI"}, {"id": 802856129, "item_id": 50178508, "url": "https://news.ycombinator.com/item?id=50178508", "by": "jph", "date": "2026-10-21", "stance": "recommend", "quote": "Claude Fable/Opus/Sonnet together successfully refactored the whole Vix IDE codebase", "excerpt": "I'm doing this attack surface tactic with Vix IDE, a terminal text editor that helps novices get comfortable with vi, emacs, git, jj, sql, commands, and more. The prompt I used was \"Goal: rework to reduce the attack surface. Process: move features into feature-flag Rust crates. Output: build a minimal version. Questions: Ask me anything.\" https://github.com/vixide/vix This past weekend, Claude Fable/Opus/Sonnet toget", "context": "", "use_case": "refactoring codebase into crates", "switched_from": "", "story": "Using Claude to reduce the attack surface of code"}, {"id": 803427424, "item_id": 50214214, "url": "https://news.ycombinator.com/item?id=50214214", "by": "throw-qqqqq", "date": "2026-10-23", "stance": "recommend", "quote": "I use Claude every day and am amazed at the quality.", "excerpt": "You have a good point. To me, this says something about the level of expertise they offer. I\u2019m a senior developer, I use Claude every day and am amazed at the quality. But I also see the shit that comes out of Claude when some colleagues, who were poor engineers to begin with, drive the AI. It\u2019s not just about the model IME. How it is used, really makes a difference. I think this is a big part of the answer: the huma", "context": "senior developer", "use_case": "software development", "switched_from": "", "story": "Ask HN: If the models are so good, why doesn't OpenAI just start 100 companies?"}]}, {"text": " reverse engineering old software and hardware,", "cites": [{"id": 805022192, "item_id": 50313887, "url": "https://news.ycombinator.com/item?id=50313887", "by": "bendangelo", "date": "2026-10-31", "stance": "recommend", "quote": "I used claude to reverse engineer an old foxpro program.", "excerpt": "I used claude to reverse engineer an old foxpro program. it setup a dos box and went through every screen. Looked at whatever files it could. Took a single week and it was 99% the same.", "context": "", "use_case": "reverse engineering legacy FoxPro program", "switched_from": "", "story": "Cisco gives engineer unlimited AI tokens to rewrite legacy code"}, {"id": 801698448, "item_id": 50106153, "url": "https://news.ycombinator.com/item?id=50106153", "by": "revolvingthrow", "date": "2026-10-15", "stance": "recommend", "quote": "I used Claude for the drivers and was essentially used by it as a meat sock to diagnose what's happening.", "excerpt": "I reverse engineered almost all electronics I own. Does it have USB that does more than just charging? Connect it to a computer and let AI play with it. Unless it's heavily secured like a phone that's pretty much all you need to do. Making some old remote work with various devices? Simple. Dumping the entire state of Rodecaster and controlling it in any you want with software, or maybe a remote? Why not. Old printer ", "context": "reverse engineers own electronics", "use_case": "writing printer drivers, reverse engineering", "switched_from": "", "story": "Using AI to reverse engineer hardware"}, {"id": 801673632, "item_id": 50104602, "url": "https://news.ycombinator.com/item?id=50104602", "by": "rvz", "date": "2026-10-15", "stance": "recommend", "quote": "now Claude can reverse engineer and build a driver to get Nvidia GPUs working on Intel Macs", "excerpt": "Instead of targeting extremely ancient hardware that almost no-one uses, what if you got Claude to reverse engineer useful closed source hardware and software? Both Apple and Nvidia are the most hostile and have the most locked-down heavy hardware and software and now Claude can reverse engineer and build a driver to get Nvidia GPUs working on Intel Macs [0] and it can unlock the ability to pair an Apple watch with A", "context": "", "use_case": "reverse engineering drivers", "switched_from": "", "story": "Using AI to reverse engineer hardware"}]}, {"text": " and diagnosing problems from logs.", "cites": [{"id": 802976624, "item_id": 50186039, "url": "https://news.ycombinator.com/item?id=50186039", "by": "leptons", "date": "2026-10-21", "stance": "recommend", "quote": "Claude has also been helping a lot with Windows, sifting through voluminous system logs to diagnose and fix obscure problems.", "excerpt": "Claude has also been helping a lot with Windows, sifting through voluminous system logs to diagnose and fix obscure problems. It's rather amazing at that stuff, it's like an all-knowing sysadmin, you just have to know what to ask and give it the logs and it tells you what to try. It's correct at least 95% of the time. My Windows system started having problems resuming from hibernate, it would just end up restarting w", "context": "Windows user since v3.0, pays $20/mo", "use_case": "diagnosing system problems from logs", "switched_from": "", "story": "Plex vulnerabilities being exploited in the wild to drop a cryptominer"}, {"id": 804897344, "item_id": 50306084, "url": "https://news.ycombinator.com/item?id=50306084", "by": "a012", "date": "2026-10-30", "stance": "recommend", "quote": "We use AI at work, Claude specifically, and it\u2019s faster than avg SRE for trivial tasks like cross signals correlation.", "excerpt": "We use AI at work, Claude specifically, and it\u2019s faster than avg SRE for trivial tasks like cross signals correlation. But it requires a lot of effort to build a harness from your company contexts for it to work like an SRE. And it\u2019s just a tool so if you have to build your knowledge to make it useful, otherwise it\u2019s just a blunt. Or in other word, incompetent people are incompetent no matter of tools", "context": "uses at work on SRE tasks", "use_case": "SRE tasks, signal correlation", "switched_from": "", "story": "AI isn't replacing SREs, it's deskilling them"}]}, {"text": " The most common complaints are refusals on security research tasks", "cites": [{"id": 801611680, "item_id": 50100730, "url": "https://news.ycombinator.com/item?id=50100730", "by": "wereHamster", "date": "2026-10-15", "stance": "warn", "quote": "It refused, so I told it", "excerpt": "Last week I got a report about a vuln in one of our products. Asked Claude to confirm it. It refused, so I told it \"fuck you, I'm switching to a chinese model to help me out here\". It replied, I'm paraphrasing, that it's \"not offended, but doesn't like the tone of voice I used towards it, and still refuses to serve my request\". Fine, so be it. I sent ^D^D back to it, switched to deepseek-v4.1-flash from within openco", "context": "maintains a product receiving vulnerability reports", "use_case": "confirming vulnerability reports", "switched_from": "", "story": "Wanted to download an ISO, ended up writing a kernel exploit"}, {"id": 803044656, "item_id": 50190291, "url": "https://news.ycombinator.com/item?id=50190291", "by": "jadar", "date": "2026-10-22", "stance": "warn", "quote": "daddy Dario won\u2019t let me because \u201ccyber\u201d", "excerpt": "I am so annoyed\u2026 I am trying to verify the claims of this article with my friend Claude but daddy Dario won\u2019t let me because \u201ccyber\u201d. I pointed it at this article and a PoC [1] I found on DuckDuckGo. We live in a stupid world. I can\u2019t run it right now so I am trying to figure out if it\u2019s serious. [1] https://github.com/eness/php2shell-CVE-2026-109618-CVE-2026-109619-PoC", "context": "", "use_case": "verifying exploit claims", "switched_from": "", "story": "There isn't an easy exploit for this PHP vulnerability"}]}, {"text": " and tight usage limits on the $20 plan.", "cites": [{"id": 804106352, "item_id": 50256647, "url": "https://news.ycombinator.com/item?id=50256647", "by": "dizhn", "date": "2026-10-27", "stance": "warn", "quote": "I tried the Claude $20 plan for two weeks and burned the rest of it due to how much token anxiety it gave me.", "excerpt": "I tried the Claude $20 plan for two weeks and burned the rest of it due to how much token anxiety it gave me. And I'm not ready to keep throwing money at it for it to become usable. Gemini models are actually fine for me as an amateur+ I found especially since they have regional pricing and a very generous side items package. I use the models but not the \"harness\" though. Not directly anyway. I am still handicapped b", "context": "amateur+", "use_case": "AI coding assistant", "switched_from": "", "story": "Anthropic S-1"}, {"id": 802518402, "item_id": 50157400, "url": "https://news.ycombinator.com/item?id=50157400", "by": "RussianCow", "date": "2026-10-19", "stance": "warn", "quote": "using Claude at API prices is prohibitively expensive.", "excerpt": "> Claude sub has a 60x subsidy, so a $20 sub gives you $1200 in API. This is only true if you max out the plan, which requires working around awkward 5-hour and weekly limits. For side projects, sure, you can use a Ralph loop or whatever and just continue when the window resets. For work, I can't afford to simply stop working when I hit my limit. And using Claude at API prices is prohibitively expensive. Personally, ", "context": "", "use_case": "work coding", "switched_from": "", "story": "MiniMax M3.1 Flash"}]}], "praise": [{"label": "Code quality", "summary": "Several developers say it writes, ports and refactors code well, and that it follows codebase conventions.", "quotes": [{"id": 804166944, "item_id": 50260434, "url": "https://news.ycombinator.com/item?id=50260434", "by": "misja111", "date": "2026-10-27", "stance": "recommend", "quote": "Claude has gotten so good that I have to admit it's a better programmer than myself.", "excerpt": "Until half a year ago I used to agree with you, but nowadays Claude has gotten so good that I have to admit it's a better programmer than myself. Its coding style is excellent, it follows whatever conventions are used in the codebase, it has excellent suggestions for refactorings and it spots omissions or mistakes in the specifications I give it. (I'm using mostly Scala as a language in a large microservices environm", "context": "works in a large microservices environment", "use_case": "programming in Scala microservices", "switched_from": "", "story": "The Mythical Man-Month in the Age of AI"}, {"id": 802856129, "item_id": 50178508, "url": "https://news.ycombinator.com/item?id=50178508", "by": "jph", "date": "2026-10-21", "stance": "recommend", "quote": "Claude Fable/Opus/Sonnet together successfully refactored the whole Vix IDE codebase", "excerpt": "I'm doing this attack surface tactic with Vix IDE, a terminal text editor that helps novices get comfortable with vi, emacs, git, jj, sql, commands, and more. The prompt I used was \"Goal: rework to reduce the attack surface. Process: move features into feature-flag Rust crates. Output: build a minimal version. Questions: Ask me anything.\" https://github.com/vixide/vix This past weekend, Claude Fable/Opus/Sonnet toget", "context": "", "use_case": "refactoring codebase into crates", "switched_from": "", "story": "Using Claude to reduce the attack surface of code"}, {"id": 803822369, "item_id": 50238898, "url": "https://news.ycombinator.com/item?id=50238898", "by": "ct520", "date": "2026-10-26", "stance": "recommend", "quote": "Claude made short work of porting to Hugo", "excerpt": "Same we moved from Wordpress to Hugo when looking to update our company site page. Using AI to update the legacy Wordpress theme just wasn\u2019t it. Had updated brand and style guide, Claude made short work of porting to Hugo and Google speed rankings were excellent even without cdn etc", "context": "", "use_case": "porting site from WordPress", "switched_from": "", "story": "Removing Next.js Got My Blog to 100 on Every Lighthouse Category"}]}, {"label": "Reverse engineering", "summary": "Used to reverse engineer old programs, devices and drivers, and to bring up firmware.", "quotes": [{"id": 805022192, "item_id": 50313887, "url": "https://news.ycombinator.com/item?id=50313887", "by": "bendangelo", "date": "2026-10-31", "stance": "recommend", "quote": "I used claude to reverse engineer an old foxpro program.", "excerpt": "I used claude to reverse engineer an old foxpro program. it setup a dos box and went through every screen. Looked at whatever files it could. Took a single week and it was 99% the same.", "context": "", "use_case": "reverse engineering legacy FoxPro program", "switched_from": "", "story": "Cisco gives engineer unlimited AI tokens to rewrite legacy code"}, {"id": 801698448, "item_id": 50106153, "url": "https://news.ycombinator.com/item?id=50106153", "by": "revolvingthrow", "date": "2026-10-15", "stance": "recommend", "quote": "I used Claude for the drivers and was essentially used by it as a meat sock to diagnose what's happening.", "excerpt": "I reverse engineered almost all electronics I own. Does it have USB that does more than just charging? Connect it to a computer and let AI play with it. Unless it's heavily secured like a phone that's pretty much all you need to do. Making some old remote work with various devices? Simple. Dumping the entire state of Rodecaster and controlling it in any you want with software, or maybe a remote? Why not. Old printer ", "context": "reverse engineers own electronics", "use_case": "writing printer drivers, reverse engineering", "switched_from": "", "story": "Using AI to reverse engineer hardware"}, {"id": 801893729, "item_id": 50118358, "url": "https://news.ycombinator.com/item?id=50118358", "by": "ckemere", "date": "2026-10-16", "stance": "recommend", "quote": "in addition to the PCB, Claude brought up the firmware and driver.", "excerpt": "I\u2019ve had really good results recently with Fable 5.1. The typical process that it uses for routing is to pin a few key traces (using the kicad python api) and then use freerouting for the others. I\u2019ve also noticed that if I push hard for \u201ccarefully designed\u201d routes, eg for a 38 pin connector, it will write code and make much better looking results. I\u2019d say that at the moment it still requires you to look at its resul", "context": "new to ST MCUs", "use_case": "firmware and driver development", "switched_from": "", "story": "AI-assisted PCB design for KiCad 10"}]}, {"label": "Debugging", "summary": "Helps diagnose problems from system logs, reproduce bugs and correlate signals.", "quotes": [{"id": 802976624, "item_id": 50186039, "url": "https://news.ycombinator.com/item?id=50186039", "by": "leptons", "date": "2026-10-21", "stance": "recommend", "quote": "Claude has also been helping a lot with Windows, sifting through voluminous system logs to diagnose and fix obscure problems.", "excerpt": "Claude has also been helping a lot with Windows, sifting through voluminous system logs to diagnose and fix obscure problems. It's rather amazing at that stuff, it's like an all-knowing sysadmin, you just have to know what to ask and give it the logs and it tells you what to try. It's correct at least 95% of the time. My Windows system started having problems resuming from hibernate, it would just end up restarting w", "context": "Windows user since v3.0, pays $20/mo", "use_case": "diagnosing system problems from logs", "switched_from": "", "story": "Plex vulnerabilities being exploited in the wild to drop a cryptominer"}, {"id": 802668720, "item_id": 50166795, "url": "https://news.ycombinator.com/item?id=50166795", "by": "perlgeek", "date": "2026-10-20", "stance": "recommend", "quote": "I asked Claude to write me a test harness that reproduced the issue, and it did.", "excerpt": "My attempts at finding a middle ground: * I use LLMs to resolve the more annoying merge conflicts. It's not super hard work, but I don't like it, and I don't think I learn much from doing it. * I implement a feature by hand, write some basic tests, and tell the AI to review it and write tests for cases that I missed (or was too lazy to do, most of the time) * For 80% of the bugs that I fix, I have a pretty good idea ", "context": "software engineer", "use_case": "writing test harnesses", "switched_from": "", "story": "Ask HN: How to deal with shipping AI-written code you don't understand?"}, {"id": 804897344, "item_id": 50306084, "url": "https://news.ycombinator.com/item?id=50306084", "by": "a012", "date": "2026-10-30", "stance": "recommend", "quote": "We use AI at work, Claude specifically, and it\u2019s faster than avg SRE for trivial tasks like cross signals correlation.", "excerpt": "We use AI at work, Claude specifically, and it\u2019s faster than avg SRE for trivial tasks like cross signals correlation. But it requires a lot of effort to build a harness from your company contexts for it to work like an SRE. And it\u2019s just a tool so if you have to build your knowledge to make it useful, otherwise it\u2019s just a blunt. Or in other word, incompetent people are incompetent no matter of tools", "context": "uses at work on SRE tasks", "use_case": "SRE tasks, signal correlation", "switched_from": "", "story": "AI isn't replacing SREs, it's deskilling them"}]}, {"label": "Value", "summary": "Some find the subscription goes much further than competitors or running models locally.", "quotes": [{"id": 803443953, "item_id": 50215247, "url": "https://news.ycombinator.com/item?id=50215247", "by": "ricardobeat", "date": "2026-10-23", "stance": "recommend", "quote": "The same $20 goes 100x further with Claude.", "excerpt": "I had the lowest plan ($20/month I think) and it literally gave me fifteen, maybe twenty minutes tops, using the cheaper K3-256K variant. Basically enough to fill the context about 1.2x and that was it. It's a really smart model but way too expensive. The same $20 goes 100x further with Claude. I used it for the initial task planning while the plan didn't expire, then handed off to DeepSeek or Minimax.", "context": "", "use_case": "coding", "switched_from": "", "story": "Kimi K3.1 Pro and K3.1 Pro Highspeed"}, {"id": 804192353, "item_id": 50262022, "url": "https://news.ycombinator.com/item?id=50262022", "by": "plqbfbv", "date": "2026-10-27", "stance": "recommend", "quote": "a 20$/mo Claude subscription is way cheaper/more efficient if you code 2-3h in the evening than running stuff locally", "excerpt": "I've benchmarked quite a few models on my 9070XT, and I basically gave up hope of using models locally: https://claude.ai/artifact/77pHgucqQNsmic8ZEnqdBA I asked Claude to come up with 7 problems to solve that were not part of any benchmarks. Besides speed, I asked it to put together a scoring that would reflect models that perform well on the 7 problems and in a reasonable time-frame (RWS, the definition is in the a", "context": "codes 2-3h in the evening", "use_case": "coding", "switched_from": "", "story": "New iMac with M6"}, {"id": 802676016, "item_id": 50167251, "url": "https://news.ycombinator.com/item?id=50167251", "by": "prologic", "date": "2026-10-20", "stance": "recommend", "quote": "Just buy the Claude Prox 5x plan. it works great.", "excerpt": "Honestly. Just buy the Claude Prox 5x plan. it works great.", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Ask HN: Best value/money AI coding plan bundle?"}]}, {"label": "Data and scripts", "summary": "Turns data dumps into dashboards and builds repeatable automation scripts.", "quotes": [{"id": 805091504, "item_id": 50318219, "url": "https://news.ycombinator.com/item?id=50318219", "by": "al_borland", "date": "2026-10-31", "stance": "recommend", "quote": "The other day I used Claude to turn a massive csv dump of data into a dashboard to see trends over time.", "excerpt": "I think a chatbot or voice is a horrible interface for most things. There is 0 chance for discoverability, so the application's function is limited by the user's imagination. Most people aren't going to think about the various things a program could solve for them. Even with a UI, most users only use a small fraction of the available options. Without some visual cue to let them know features exist, I'd bet that most ", "context": "works on automation in a large organization", "use_case": "turning CSV data into dashboards", "switched_from": "", "story": "Ask HN: Will apps be going away in favor of personal assistants?"}, {"id": 804818240, "item_id": 50301140, "url": "https://news.ycombinator.com/item?id=50301140", "by": "tstrimple", "date": "2026-10-30", "stance": "recommend", "quote": "I use Claude for this stuff all the time.", "excerpt": "I use Claude for this stuff all the time. The difference is I have Claude create repeatable scripts which do the actual things on a schedule. Claude is never invoked in on a schedule, only by me. One example is job board search filtered for roles I'd be interested in. The script itself includes all the things I care about like being remote. Clear salary ranges. Type of work I want to be doing. Claude built a good scr", "context": "", "use_case": "building scripts, summarizing listings", "switched_from": "", "story": "Claude Duet is the first personal-assistant-type harness that feels useful"}, {"id": 804281168, "item_id": 50267573, "url": "https://news.ycombinator.com/item?id=50267573", "by": "abirch", "date": "2026-10-27", "stance": "recommend", "quote": "Yes, Claude + skills is what we use too.", "excerpt": "Yes, Claude + skills is what we use too. Big picture, all of the old BI players are living on borrowed time. If you have to drag and drop; you'll be dropped.", "context": "", "use_case": "business intelligence / analytics", "switched_from": "Hex", "story": "Hex Raises $100M"}]}, {"label": "Learning aid", "summary": "Explains unfamiliar code and concepts while you work.", "quotes": [{"id": 803717280, "item_id": 50232330, "url": "https://news.ycombinator.com/item?id=50232330", "by": "owls-on-wires", "date": "2026-10-25", "stance": "recommend", "quote": "I just pointed Claude at the repo at asked it to explain, and it did a good job.", "excerpt": "Unfortunately no paper yet, but I\u2019m considering publishing one here soon! As far as agents go, I just pointed Claude at the repo at asked it to explain, and it did a good job. It also helps that it\u2019s not a ton of code, and a lot of it is structurally similar to patterns in training data (in this case, it can recognize a recursive beam-search). Would highly recommended submitting a PR for SRBench! The maintainers were", "context": "", "use_case": "explaining code repository", "switched_from": "", "story": "Symbolic Learning in Python and Julia"}, {"id": 802580096, "item_id": 50161256, "url": "https://news.ycombinator.com/item?id=50161256", "by": "LoganDark", "date": "2026-10-20", "stance": "recommend", "quote": "I'm working on a project with Claude right now, and it is kind of fun to have it explain stuff like HKDF to me", "excerpt": "I'm working on a project with Claude right now, and it is kind of fun to have it explain stuff like HKDF to me while I'm trying to understand its implementation. Also stuff like Ed25519 not being an encryption scheme, only signature. I love fun facts. Maybe I'm not as much vibe coding as people who only talk about the user experience without worrying about the implementation. I care deeply about all observable detail", "context": "", "use_case": "building and learning crypto project", "switched_from": "", "story": "It's still ok to enjoy programming"}, {"id": 802551217, "item_id": 50159451, "url": "https://news.ycombinator.com/item?id=50159451", "by": "Assafkip", "date": "2026-10-19", "stance": "recommend", "quote": "I use Claude to learn new stuff all the time.", "excerpt": "I use Claude to learn new stuff all the time. But reading long tutorials is a pain. So I made a skill for that. I give it a topic, it researches it, and then it makes a set consumable 2-minute videos I can listen to anytime. It's kind of like udemy and LinkedIn learning and those guys but free and for whatever you want. I made a short video for And an implementation of the repo into a simple tool. https://www.reddit.", "context": "", "use_case": "learning new topics", "switched_from": "", "story": null}]}], "caveats": [{"label": "Security refusals", "summary": "It refuses to help verify vulnerabilities and other security-flagged requests.", "quotes": [{"id": 801611680, "item_id": 50100730, "url": "https://news.ycombinator.com/item?id=50100730", "by": "wereHamster", "date": "2026-10-15", "stance": "warn", "quote": "It refused, so I told it", "excerpt": "Last week I got a report about a vuln in one of our products. Asked Claude to confirm it. It refused, so I told it \"fuck you, I'm switching to a chinese model to help me out here\". It replied, I'm paraphrasing, that it's \"not offended, but doesn't like the tone of voice I used towards it, and still refuses to serve my request\". Fine, so be it. I sent ^D^D back to it, switched to deepseek-v4.1-flash from within openco", "context": "maintains a product receiving vulnerability reports", "use_case": "confirming vulnerability reports", "switched_from": "", "story": "Wanted to download an ISO, ended up writing a kernel exploit"}, {"id": 803044656, "item_id": 50190291, "url": "https://news.ycombinator.com/item?id=50190291", "by": "jadar", "date": "2026-10-22", "stance": "warn", "quote": "daddy Dario won\u2019t let me because \u201ccyber\u201d", "excerpt": "I am so annoyed\u2026 I am trying to verify the claims of this article with my friend Claude but daddy Dario won\u2019t let me because \u201ccyber\u201d. I pointed it at this article and a PoC [1] I found on DuckDuckGo. We live in a stupid world. I can\u2019t run it right now so I am trying to figure out if it\u2019s serious. [1] https://github.com/eness/php2shell-CVE-2026-109618-CVE-2026-109619-PoC", "context": "", "use_case": "verifying exploit claims", "switched_from": "", "story": "There isn't an easy exploit for this PHP vulnerability"}, {"id": 805138048, "item_id": 50321128, "url": "https://news.ycombinator.com/item?id=50321128", "by": "silentkat", "date": "2026-10-31", "stance": "warn", "quote": "Several times it triggered this flag.", "excerpt": "I have been doing RL for Magic the Gathering (longshot but having fun). Several times it triggered this flag. But that was a couple months ago. Everyone went crazy over Astra so I switched to open ai and had no issues. People are now going crazy over opus 5.5 so I\u2019m playing with that. Haven\u2019t hit any issues so far, but two months in the code is very dirty but functional enough that I\u2019m redoing it with more heavy hand", "context": "", "use_case": "reinforcement learning code", "switched_from": "", "story": "Anthropic blocks users for anything touching training larger models"}]}, {"label": "Prose quality", "summary": "Some dislike its writing style and its answers to non-coding questions.", "quotes": [{"id": 804688018, "item_id": 50293001, "url": "https://news.ycombinator.com/item?id=50293001", "by": "novaRom", "date": "2026-10-29", "stance": "warn", "quote": "it's close to nothing when it's about non coding questions. I don't understand how anyone can like its gibberish answers?", "excerpt": "I have switched to Mistral free app just from US apps completely recently - it beats for me absolutely all alternatives: ChatGPT - stopped using it like a year ago, it feels they deliberately made it ridiculously stupid (for fun?). I know, they were first, but today it's a joke. Claude - tried it out of curiosity few times, but it's close to nothing when it's about non coding questions. I don't understand how anyone ", "context": "based in EU", "use_case": "non-coding questions", "switched_from": "", "story": "We're releasing the weights of Mistral Large 4"}, {"id": 803429936, "item_id": 50214371, "url": "https://news.ycombinator.com/item?id=50214371", "by": "KronisLV", "date": "2026-10-23", "stance": "warn", "quote": "I used Claude a lot and absolutely hated it", "excerpt": "I used Claude a lot and absolutely hated it: https://blog.kronis.dev/blog/ai-slop-is-a-self-inflicted-tragedy/ Though to be fair Opus 5.5 is better than Opus 5 was in that regard. I still prefer the OpenAI models to either of them when it comes to regular prose, though.", "context": "", "use_case": "writing and coding", "switched_from": "", "story": "Kimi K3.1 Pro and K3.1 Pro Highspeed"}, {"id": 802800352, "item_id": 50175022, "url": "https://news.ycombinator.com/item?id=50175022", "by": "nunez", "date": "2026-10-21", "stance": "warn", "quote": "Claude's EXTREMELY tired \"we have Axios at home\" headlines", "excerpt": "I love the closing artwork. It's really on theme. I agree with the author. Content marketers, journalists and other professional writers are very good at making content sound engaging without losing the \"voice\" of their clients. Tech writers are generally good at presenting complex topics in very accessible ways. They develop style guides that make this easier to do at scale, and editors who are usually better at rea", "context": "", "use_case": "writing headlines", "switched_from": "", "story": "On AI writing"}]}, {"label": "Flawed code", "summary": "It can produce poor or uncompilable code in some libraries and languages, and it struggled with KiCad schematics.", "quotes": [{"id": 803714560, "item_id": 50232160, "url": "https://news.ycombinator.com/item?id=50232160", "by": "munro", "date": "2026-10-25", "stance": "warn", "quote": "It said code generate with Claude, is it really still this bad?", "excerpt": "IDK lot of noise in report, I need more signal to noise ratio, already reading so much AI output as it is. But the example PR... It said code generate with Claude, is it really still this bad? Last I was using SQLAlchemy with Claude was Opus 4.5, and it took some heavy coaching and skill gen not to generate code like this--but why even use SQLAlchemy at all if you're just going to use string interpolation jeez https:", "context": "", "use_case": "generating SQLAlchemy code", "switched_from": "", "story": "Show HN: Farol \u2013 A risk map of a pull request for the person who approves it"}, {"id": 802356161, "item_id": 50147260, "url": "https://news.ycombinator.com/item?id=50147260", "by": "Perenti", "date": "2026-10-19", "stance": "warn", "quote": "all the models seem to love writing uncompileable code, and ending up in fearsome knots over it", "excerpt": "I love Free Pascal. I've tried using claude/codex/qwen to write in it, but all the models seem to love writing uncompileable code, and ending up in fearsome knots over it. It's my number one \"knock something together that just works\" language for humans though.", "context": "", "use_case": "writing Free Pascal code", "switched_from": "", "story": "Free Pascal compiler now supports dynamic packages"}, {"id": 801892704, "item_id": 50118294, "url": "https://news.ycombinator.com/item?id=50118294", "by": "the__alchemist", "date": "2026-10-16", "stance": "warn", "quote": "tried to get Claude and COdex to do the schematic in KiCad, and they were not effective.", "excerpt": "I will give this a try! I feel like AI should be able to automate this process (Schematic and/or PCB). I made a new design 2 months ago; tried to get Claude and COdex to do the schematic in KiCad, and they were not effective. Made a mess of it for an initial build, and edits later. I didn't bother with the component placement and routing. I did find LLMs to be very effective on answering questions about routing, part", "context": "embedded developer using Rust", "use_case": "KiCad schematic generation", "switched_from": "", "story": "AI-assisted PCB design for KiCad 10"}]}, {"label": "Usage limits", "summary": "The $20 plan's limits cause token anxiety, and API pricing is called prohibitively expensive.", "quotes": [{"id": 804106352, "item_id": 50256647, "url": "https://news.ycombinator.com/item?id=50256647", "by": "dizhn", "date": "2026-10-27", "stance": "warn", "quote": "I tried the Claude $20 plan for two weeks and burned the rest of it due to how much token anxiety it gave me.", "excerpt": "I tried the Claude $20 plan for two weeks and burned the rest of it due to how much token anxiety it gave me. And I'm not ready to keep throwing money at it for it to become usable. Gemini models are actually fine for me as an amateur+ I found especially since they have regional pricing and a very generous side items package. I use the models but not the \"harness\" though. Not directly anyway. I am still handicapped b", "context": "amateur+", "use_case": "AI coding assistant", "switched_from": "", "story": "Anthropic S-1"}, {"id": 802518402, "item_id": 50157400, "url": "https://news.ycombinator.com/item?id=50157400", "by": "RussianCow", "date": "2026-10-19", "stance": "warn", "quote": "using Claude at API prices is prohibitively expensive.", "excerpt": "> Claude sub has a 60x subsidy, so a $20 sub gives you $1200 in API. This is only true if you max out the plan, which requires working around awkward 5-hour and weekly limits. For side projects, sure, you can use a Ralph loop or whatever and just continue when the window resets. For work, I can't afford to simply stop working when I hit my limit. And using Claude at API prices is prohibitively expensive. Personally, ", "context": "", "use_case": "work coding", "switched_from": "", "story": "MiniMax M3.1 Flash"}]}, {"label": "App reliability", "summary": "Users report a buggy app and failed integrations, such as with Google Docs.", "quotes": [{"id": 802998240, "item_id": 50187390, "url": "https://news.ycombinator.com/item?id=50187390", "by": "flakiness", "date": "2026-10-21", "stance": "recommend", "quote": "I love their model but Meta/Muse has the edge on non-model implementation.", "excerpt": "I worry how buggy their app is. I love their model but Meta/Muse has the edge on non-model implementation. Hope this turn out to be false alarm.", "context": "", "use_case": "", "switched_from": "", "story": "Claude Duet"}, {"id": 803385682, "item_id": 50211605, "url": "https://news.ycombinator.com/item?id=50211605", "by": "joshstrange", "date": "2026-10-23", "stance": "warn", "quote": "login to Claude in the new sidebar, and.... \"I can't read the document\"", "excerpt": "And it doesn't work if you are signed into multiple Google accounts :/ Today I wanted to fix some mixed formatting in a Google Doc. Ok, I'll use Gemini, it's right there and surely it's the best. Nope, fails twice \"Please try again\" then suggests just trash formatting. Ok, fine, I connected this to ChatGPT, let's try that. Nope, multiple rounds of \"I can't see that document\" despite having the plugin configured. Ok, ", "context": "signed into multiple Google accounts", "use_case": "fixing Google Doc formatting", "switched_from": "", "story": "Claude Now Works with Google Docs, Sheets, and Slides"}]}], "who": [{"by": "al_borland", "context": "works on automation in a large organization", "url": "https://news.ycombinator.com/item?id=50318219", "date": "2026-10-31"}, {"by": "Jash271", "context": "not a video editor", "url": "https://news.ycombinator.com/item?id=50311807", "date": "2026-10-31"}, {"by": "js8", "context": "Max 5x subscriber", "url": "https://news.ycombinator.com/item?id=50307038", "date": "2026-10-30"}, {"by": "a012", "context": "uses at work on SRE tasks", "url": "https://news.ycombinator.com/item?id=50306084", "date": "2026-10-30"}, {"by": "martypitt", "context": "startup founder with investors", "url": "https://news.ycombinator.com/item?id=50301583", "date": "2026-10-30"}, {"by": "bradknowles", "context": "technical writer", "url": "https://news.ycombinator.com/item?id=50298961", "date": "2026-10-30"}, {"by": "hoppp", "context": "solo, no money to hire people", "url": "https://news.ycombinator.com/item?id=50287402", "date": "2026-10-29"}, {"by": "munch-o-man", "context": "developer who reviews all output", "url": "https://news.ycombinator.com/item?id=50285238", "date": "2026-10-29"}, {"by": "jaaron", "context": "Emacs user", "url": "https://news.ycombinator.com/item?id=50281054", "date": "2026-10-28"}, {"by": "waiwai933", "context": "US citizen living abroad preparing own taxes", "url": "https://news.ycombinator.com/item?id=50276596", "date": "2026-10-28"}, {"by": "plqbfbv", "context": "codes 2-3h in the evening", "url": "https://news.ycombinator.com/item?id=50262022", "date": "2026-10-27"}, {"by": "misja111", "context": "works in a large microservices environment", "url": "https://news.ycombinator.com/item?id=50260434", "date": "2026-10-27"}], "tags": ["for Emacs users", "for homelab automation", "for reverse engineering", "for SRE work", "for technical writing"], "use_cases": ["coding", "ai coding", "ai assistant", "coding agent", "writing code", "work"], "compared": [{"slug": "codex", "name": "Codex", "co": 306, "switched_from": 2, "switched_to": 47}, {"slug": "chatgpt", "name": "ChatGPT", "co": 195, "switched_from": 12, "switched_to": 9}, {"slug": "gemini", "name": "Gemini", "co": 146, "switched_from": 1, "switched_to": 3}, {"slug": "deepseek", "name": "DeepSeek", "co": 69, "switched_from": 0, "switched_to": 14}, {"slug": "glm", "name": "GLM", "co": 56, "switched_from": 0, "switched_to": 14}, {"slug": "cursor", "name": "Cursor", "co": 60, "switched_from": 1, "switched_to": 0}, {"slug": "opencode", "name": "OpenCode", "co": 50, "switched_from": 0, "switched_to": 6}, {"slug": "gpt", "name": "GPT", "co": 51, "switched_from": 1, "switched_to": 3}], "makers": [{"id": 793408544, "item_id": 49588034, "url": "https://news.ycombinator.com/item?id=49588034", "by": "bcherny", "date": "2026-09-06", "stance": "recommend", "quote": "it's just Claude working on the code", "excerpt": "(I work at Anthropic) I agree. I use an LLM to write my code, but I do all of my writing by hand, since it helps me think. It reminds me of the transition over the last year from AI-assisted coding to AI doing all the coding. At first the code output wasn't good enough, and humans read and iterated on the code all day, so the details of the source code mattered. Now, the code is largely high quality and it meets a la", "context": "works at Anthropic", "use_case": "writing code", "switched_from": "", "story": "Your intellectual fly is open when you use an LLM to author a post (2025)"}], "threads": [{"url": "https://news.ycombinator.com/item?id=48449187", "title": "Ask HN: What are tools you have made for yourself since the advent of AI", "mentions": 17}, {"url": "https://news.ycombinator.com/item?id=49672966", "title": "Ask HN: What default model do you use and why?", "mentions": 12}, {"url": "https://news.ycombinator.com/item?id=48406174", "title": "What was your \"Oh Shit\" moment with GenAI?", "mentions": 10}, {"url": "https://news.ycombinator.com/item?id=49375996", "title": "Clean up Claude 5's token vomit with a separate LLM", "mentions": 8}, {"url": "https://news.ycombinator.com/item?id=49151933", "title": "Don't be a meat proxy", "mentions": 7}, {"url": "https://news.ycombinator.com/item?id=48849066", "title": "GPT-5.6", "mentions": 7}], "sentiment_by_month": {"2025-11": {"compare": 3, "neutral": 17, "recommend": 16, "warn": 4}, "2025-12": {"recommend": 25, "warn": 4, "neutral": 15}, "2026-01": {"neutral": 30, "recommend": 33, "warn": 6, "compare": 3}, "2026-02": {"recommend": 39, "neutral": 27, "warn": 6, "compare": 6}, "2026-03": {"recommend": 65, "neutral": 42, "warn": 11, "compare": 6}, "2026-04": {"recommend": 36, "compare": 3, "neutral": 36, "warn": 23}, "2026-05": {"recommend": 49, "neutral": 45, "compare": 7, "warn": 7}, "2026-06": {"recommend": 93, "compare": 16, "neutral": 95, "warn": 22}, "2026-07": {"compare": 9, "neutral": 90, "recommend": 90, "warn": 41}, "2026-08": {"neutral": 94, "warn": 50, "recommend": 91, "compare": 6}, "2026-09": {"neutral": 101, "recommend": 71, "warn": 44, "compare": 13}, "2026-10": {"neutral": 92, "recommend": 86, "compare": 4, "warn": 18}}, "quotes": [{"id": 805138497, "item_id": 50321156, "url": "https://news.ycombinator.com/item?id=50321156", "by": "NorthSouthNorth", "date": "2026-10-31", "stance": "recommend", "quote": "I prefer Claude's current model and their native web app over that of OpenAI.", "excerpt": "Ja, that's the only reason I use ChatGPT instead even though I prefer Claude's current model and their native web app over that of OpenAI.", "context": "", "use_case": "AI assistant", "switched_from": "", "story": "Prospero: A new coding harness for a more civilized age"}, {"id": 805138048, "item_id": 50321128, "url": "https://news.ycombinator.com/item?id=50321128", "by": "silentkat", "date": "2026-10-31", "stance": "warn", "quote": "Several times it triggered this flag.", "excerpt": "I have been doing RL for Magic the Gathering (longshot but having fun). Several times it triggered this flag. But that was a couple months ago. Everyone went crazy over Astra so I switched to open ai and had no issues. People are now going crazy over opus 5.5 so I\u2019m playing with that. Haven\u2019t hit any issues so far, but two months in the code is very dirty but functional enough that I\u2019m redoing it with more heavy hand", "context": "", "use_case": "reinforcement learning code", "switched_from": "", "story": "Anthropic blocks users for anything touching training larger models"}, {"id": 805105056, "item_id": 50319066, "url": "https://news.ycombinator.com/item?id=50319066", "by": "bfbf", "date": "2026-10-31", "stance": "neutral", "quote": "", "excerpt": "On a micro scale, yes I see this regularly at work: some devs blindly put up ai generated PRs that are fundamentally flawed or don\u2019t actually complete the work in the ticket. It\u2019s often trivial to check this. Similarly some tickets are written by ai that make incorrect assumptions about our product, resulting in nonsensical requirements. You could argue that these are just examples of laziness, but I think these are ", "context": "", "use_case": "evaluating SDKs", "switched_from": "", "story": "The Impact of Generative AI on Critical Thinking: Self-Reported (2025)"}, {"id": 805091504, "item_id": 50318219, "url": "https://news.ycombinator.com/item?id=50318219", "by": "al_borland", "date": "2026-10-31", "stance": "recommend", "quote": "The other day I used Claude to turn a massive csv dump of data into a dashboard to see trends over time.", "excerpt": "I think a chatbot or voice is a horrible interface for most things. There is 0 chance for discoverability, so the application's function is limited by the user's imagination. Most people aren't going to think about the various things a program could solve for them. Even with a UI, most users only use a small fraction of the available options. Without some visual cue to let them know features exist, I'd bet that most ", "context": "works on automation in a large organization", "use_case": "turning CSV data into dashboards", "switched_from": "", "story": "Ask HN: Will apps be going away in favor of personal assistants?"}, {"id": 805022192, "item_id": 50313887, "url": "https://news.ycombinator.com/item?id=50313887", "by": "bendangelo", "date": "2026-10-31", "stance": "recommend", "quote": "I used claude to reverse engineer an old foxpro program.", "excerpt": "I used claude to reverse engineer an old foxpro program. it setup a dos box and went through every screen. Looked at whatever files it could. Took a single week and it was 99% the same.", "context": "", "use_case": "reverse engineering legacy FoxPro program", "switched_from": "", "story": "Cisco gives engineer unlimited AI tokens to rewrite legacy code"}, {"id": 804988912, "item_id": 50311807, "url": "https://news.ycombinator.com/item?id=50311807", "by": "Jash271", "date": "2026-10-31", "stance": "neutral", "quote": "I used Claude to generate the videos", "excerpt": "Agents interacting with each other made for great content, so we created a series out of it: https://www.youtube.com/watch?v=dxXvgM75yis PS: I'm not a video editor. I used Claude to generate the videos, ElevenLabs for the voice, and GPT for the images of the characters", "context": "not a video editor", "use_case": "generating videos", "switched_from": "", "story": "Where AI agents' social boundaries break: 5 LLMs try to form a government"}, {"id": 804949568, "item_id": 50309348, "url": "https://news.ycombinator.com/item?id=50309348", "by": "giancarlostoro", "date": "2026-10-30", "stance": "recommend", "quote": "I love using Claude", "excerpt": "It's the cryptobro craze but for AI. It's tiresome, I don't have AI fatigue, I love using Claude, I have aibro fatigue. I'll lump in the people all in on AI, and that includes anyone who hates it vehemently because they just have to sour conversations.", "context": "", "use_case": "", "switched_from": "", "story": "Anthropic may not secure investment-grade rating after IPO, Blackstone says"}, {"id": 804912608, "item_id": 50307038, "url": "https://news.ycombinator.com/item?id=50307038", "by": "js8", "date": "2026-10-30", "stance": "neutral", "quote": "I asked Claude and it estimated 10-50 kWh if you max out Max 5x subscription (which I have", "excerpt": "Yes, I asked Claude and it estimated 10-50 kWh if you max out Max 5x subscription (which I have, and I use it about 30-40%). Further estimation gave it's about like having another gaming PC, i.e. 100-300W when I am actually gaming. I live quite low carbon intensity lifestyle, don't travel much, don't have a car or a family, so it's not really a big expense.", "context": "Max 5x subscriber", "use_case": "AI assistant", "switched_from": "", "story": "We ran AI agent swarms in Minecraft and Mindustry"}, {"id": 804897344, "item_id": 50306084, "url": "https://news.ycombinator.com/item?id=50306084", "by": "a012", "date": "2026-10-30", "stance": "recommend", "quote": "We use AI at work, Claude specifically, and it\u2019s faster than avg SRE for trivial tasks like cross signals correlation.", "excerpt": "We use AI at work, Claude specifically, and it\u2019s faster than avg SRE for trivial tasks like cross signals correlation. But it requires a lot of effort to build a harness from your company contexts for it to work like an SRE. And it\u2019s just a tool so if you have to build your knowledge to make it useful, otherwise it\u2019s just a blunt. Or in other word, incompetent people are incompetent no matter of tools", "context": "uses at work on SRE tasks", "use_case": "SRE tasks, signal correlation", "switched_from": "", "story": "AI isn't replacing SREs, it's deskilling them"}, {"id": 804854657, "item_id": 50303416, "url": "https://news.ycombinator.com/item?id=50303416", "by": "darkwi11ow", "date": "2026-10-30", "stance": "neutral", "quote": "I use Claude, Codex and Cursor/SpaceX agents", "excerpt": "Well for me, as developer on a modern stack, there is always some risk involved. If not from agentic tools, then the package registries are compromised left and right, so it all comes down to mitigating the risk. I use Claude, Codex and Cursor/SpaceX agents, and Cursor comes very convenient to have all three in a single interface. Also I still code manually especially when prototyping, so having an IDE helps with it ", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Backdoor in Claude Code for VS Code extension on the Cursor Marketplace"}, {"id": 804853809, "item_id": 50303363, "url": "https://news.ycombinator.com/item?id=50303363", "by": "martincsweiss", "date": "2026-10-30", "stance": "neutral", "quote": "I used Claude and Codex to label the symbols", "excerpt": "Author here, and Postdoc with Hugo Larochelle and Chris Pal. I saw the LocalWiki last week, and it inspired me to make a learning tool based on \"local\" research papers. Many ML researchers write code that relies on a long history of research in ML and applied math. But which papers are you actually building on? You can paste a GitHub repo and Standing on will read its Python code match pieces of it back to the resear", "context": "", "use_case": "labeling symbols", "switched_from": "", "story": "Show HN: What research is your code standing on?"}, {"id": 804851745, "item_id": 50303234, "url": "https://news.ycombinator.com/item?id=50303234", "by": "darkwi11ow", "date": "2026-10-30", "stance": "neutral", "quote": "I use Claude installed via OpenVSX", "excerpt": "I only use agentic development tools in qemu VM that has no production access or host shared folders, and the only secrets there are development. I still have my Git ssh key there, but I never use ssh agent to load the keys. The worst that agentic tools can do to me, is to install keylogger and compromise my ssh key and password, but even then my deployment workflow includes creating release in Github, which is not p", "context": "", "use_case": "agentic coding", "switched_from": "", "story": "Backdoor in Claude Code for VS Code extension on the Cursor Marketplace"}, {"id": 804829808, "item_id": 50301863, "url": "https://news.ycombinator.com/item?id=50301863", "by": "zuInnp", "date": "2026-10-30", "stance": "neutral", "quote": "The fear of loosing my coding abilities because Claude does is.", "excerpt": "Before writing about how I feel about AI, some ideas that sometimes help me: - plan your breaks, e.g. \"1h hour of work and then go away from the computer for 5 or 10 minutes\" * try to break the \"just plan a little further\" loop - accept that you cannot be concentrated for hours and that reading an AI chat is mentally challenging (see first point about breaks) - Talk with your coworkers about it. You are not alone, it", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Ask HN: Are you tired too in this AI era?"}, {"id": 804829121, "item_id": 50301820, "url": "https://news.ycombinator.com/item?id=50301820", "by": "FailMore", "date": "2026-10-30", "stance": "neutral", "quote": "I would have eight Claude sessions running", "excerpt": "I went through a period of feeling very tired. For me two things helped: 1. Being okay with being \"less\" productive. I'm probably still 10x more productive than I was in the pre-AI era, but in the early days of AI I would have eight Claude sessions running. Now it's two or three and I take more advantage of the free time I have between Claude assignments. My bosses are still happy with my output and I'm not being bur", "context": "", "use_case": "AI coding assistant", "switched_from": "", "story": "Ask HN: Are you tired too in this AI era?"}, {"id": 804825328, "item_id": 50301583, "url": "https://news.ycombinator.com/item?id=50301583", "by": "martypitt", "date": "2026-10-30", "stance": "neutral", "quote": "The code Claude produces is often clean, but terribly archictected.", "excerpt": "It's weird times. I feel incredibly productive. I'm building in an afternoon what would've taken weeks. I spend all my time in either MR's reviewing code, or a terminal prompting. I've moved from being a coder, to being a product-engineer-architect. I love coding -- LOVE it. But the role has morphed. And, I actually really like the new role. I love being product centric, and caring about architecture -- it suits me w", "context": "startup founder with investors", "use_case": "AI coding", "switched_from": "", "story": "Ask HN: Are you tired too in this AI era?"}, {"id": 804818240, "item_id": 50301140, "url": "https://news.ycombinator.com/item?id=50301140", "by": "tstrimple", "date": "2026-10-30", "stance": "recommend", "quote": "I use Claude for this stuff all the time.", "excerpt": "I use Claude for this stuff all the time. The difference is I have Claude create repeatable scripts which do the actual things on a schedule. Claude is never invoked in on a schedule, only by me. One example is job board search filtered for roles I'd be interested in. The script itself includes all the things I care about like being remote. Clear salary ranges. Type of work I want to be doing. Claude built a good scr", "context": "", "use_case": "building scripts, summarizing listings", "switched_from": "", "story": "Claude Duet is the first personal-assistant-type harness that feels useful"}, {"id": 804817680, "item_id": 50301105, "url": "https://news.ycombinator.com/item?id=50301105", "by": "tstrimple", "date": "2026-10-30", "stance": "neutral", "quote": "I've built hooks and processes against it.", "excerpt": "I have literally never seen this and it doesn't make much since because the turns are flagged with metadata. It should be impossible for Claude to confuse your text for its own. I know because I use this metadata all the time. I've built hooks and processes against it.", "context": "", "use_case": "hooks and processes", "switched_from": "", "story": "Claude Duet is the first personal-assistant-type harness that feels useful"}, {"id": 804811009, "item_id": 50300688, "url": "https://news.ycombinator.com/item?id=50300688", "by": "thecopy", "date": "2026-10-30", "stance": "neutral", "quote": "You can tell Claude to adjust the config files", "excerpt": "I recommend everyone to move your plantext secrets from ~/.kube/config, ~/.ssh etc into a password-manager. You can tell Claude to adjust the config files to get the secrets from 1Password when they are needed with TouchID requirements etc.", "context": "", "use_case": "editing config files", "switched_from": "", "story": "Binance-hacker-1029 \u2013 $5.6B stolen from Binance"}, {"id": 804791696, "item_id": 50299481, "url": "https://news.ycombinator.com/item?id=50299481", "by": "UberMouse", "date": "2026-10-30", "stance": "neutral", "quote": "I use on average close to $3k a week at API prices on my 20x plan", "excerpt": ">On the 20x plan, 1% of weekly limit is worth about $20 in API prices. A full week is then about $2,000 Considering I use on average close to $3k a week at API prices on my 20x plan I don't think this quite accurate. Some weeks I've used >$3.5k", "context": "", "use_case": "heavy coding usage", "switched_from": "", "story": "Claude Code regularly sends each background agent\u2019s transcript to its model"}, {"id": 804783376, "item_id": 50298961, "url": "https://news.ycombinator.com/item?id=50298961", "by": "bradknowles", "date": "2026-10-30", "stance": "recommend", "quote": "This is exactly how I use Claude in all of my technical writing.", "excerpt": "This is exactly how I use Claude in all of my technical writing. I even wrote a skill for this. Every claim has to be backed up by a reference. Every reference has to be double-checked. Primary sources are much preferred over secondary sources, and where no primary source can be found, that must be noted. I also insist on simpler, more easy-to-understand writing, even if that makes the total output longer. Aiming at ", "context": "technical writer", "use_case": "technical writing with citations", "switched_from": "", "story": "Zombie stats: Where the numbers AI quotes come from"}, {"id": 804770705, "item_id": 50298169, "url": "https://news.ycombinator.com/item?id=50298169", "by": "drewbitt", "date": "2026-10-30", "stance": "neutral", "quote": "I use Claude too", "excerpt": "This is drastically different from how I live and work. I may do over 100 (Kagi) searches a day. I use Claude too but I am just faster searching after decades of experience.", "context": "", "use_case": "AI assistant", "switched_from": "", "story": "Alphabet heads for a $500B drop in market value, biggest in its history"}, {"id": 804762400, "item_id": 50297650, "url": "https://news.ycombinator.com/item?id=50297650", "by": "Jash271", "date": "2026-10-29", "stance": "neutral", "quote": "I used Claude to generate the videos", "excerpt": "Agents interacting with each other made for great content, so we created a series out of it: https://www.youtube.com/watch?v=dxXvgM75yis PS: I'm not a video editor. I used Claude to generate the videos, ElevenLabs for the voice, and GPT for the images of the characters.", "context": "", "use_case": "generating videos", "switched_from": "", "story": "Where AI agents' social boundaries break: 5 LLMs try to form a government"}, {"id": 804711056, "item_id": 50294441, "url": "https://news.ycombinator.com/item?id=50294441", "by": "arjie", "date": "2026-10-29", "stance": "neutral", "quote": "I suppose I read so much of this with Claude that it gets stuck in my head!", "excerpt": "The other day I was writing a blog post summarizing how I use an adaptation of a context handling technique and I wrote the following three lines: * Every user run becomes a memo * Days summarize, rolled to weeks, rolled to months - all written once * Today's runs are shortened and placed into running context past the cache break Then while reading it before publishing it, I thought it sounded super LLM-y. I didn't r", "context": "", "use_case": "", "switched_from": "", "story": "LLM Voice"}, {"id": 804700032, "item_id": 50293752, "url": "https://news.ycombinator.com/item?id=50293752", "by": "jillesvangurp", "date": "2026-10-29", "stance": "recommend", "quote": "These days, I have claude and chatgpt. They search on my behalf.", "excerpt": "Blaming AI is correct but perhaps not the way you think. I think a lot of people actually use other people's AI tools to do their searches now. Google's attempt to grab that back via their own AI tools is so far not really working. Google is no longer the go to tool for answering questions or doing research. I used perplexity for some time. These days, I have claude and chatgpt. They search on my behalf. Sometimes I ", "context": "", "use_case": "web search and research", "switched_from": "Perplexity", "story": "Google Search is now less than half of Alphabet\u2019s revenue"}, {"id": 804688018, "item_id": 50293001, "url": "https://news.ycombinator.com/item?id=50293001", "by": "novaRom", "date": "2026-10-29", "stance": "warn", "quote": "it's close to nothing when it's about non coding questions. I don't understand how anyone can like its gibberish answers?", "excerpt": "I have switched to Mistral free app just from US apps completely recently - it beats for me absolutely all alternatives: ChatGPT - stopped using it like a year ago, it feels they deliberately made it ridiculously stupid (for fun?). I know, they were first, but today it's a joke. Claude - tried it out of curiosity few times, but it's close to nothing when it's about non coding questions. I don't understand how anyone ", "context": "based in EU", "use_case": "non-coding questions", "switched_from": "", "story": "We're releasing the weights of Mistral Large 4"}, {"id": 804685952, "item_id": 50292872, "url": "https://news.ycombinator.com/item?id=50292872", "by": "nsarrazin", "date": "2026-10-29", "stance": "recommend", "quote": "I don't know - Claude did!", "excerpt": "I don't know - Claude did! For the speakers I used a Louder-ESP32-Plus board which worked great out of the box: https://sonocotta.com/louder-esp32/ For the vinyl player it's a DIY build on a breadboard and I did have to use an ESP board with an external antenna. The PCB antenna was not working great on its own.", "context": "", "use_case": "building ESP32 audio project", "switched_from": "", "story": "SIMD-accelerated computer vision on a $2 microcontroller"}, {"id": 804661616, "item_id": 50291351, "url": "https://news.ycombinator.com/item?id=50291351", "by": "rickydroll", "date": "2026-10-29", "stance": "recommend", "quote": "if I used Claude, I'd probably get addicted to the results.", "excerpt": "Good to know. I figured I'd start with Gemini because if I used Claude, I'd probably get addicted to the results. Then I would have to commission a velvet painting of me dressed in Dickensian beggar clothes with big eyes and a quivering lower lip, holding my hands out, asking for tokens.", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 4 Carbon is the new #1 on the Artificial Analysis Intelligence Index v5"}, {"id": 804634404, "item_id": 50289650, "url": "https://news.ycombinator.com/item?id=50289650", "by": "dichiara19", "date": "2026-10-29", "stance": "warn", "quote": "as often happens with Claude and OpenAI", "excerpt": "I want to talk about my experience with Argon, noting that I used it via OpenRouter on a Raspberry Pi, spending around $15 in credit. My experience is that Google is aiming to create a generalist model \u2013 but a truly generalist one! In fact, I find it particularly accurate in general terms and even in long conversations, maintaining excellent consistency and a wealth of information in response to complex questions. I\u2019", "context": "", "use_case": "supervising long agent tasks", "switched_from": "", "story": "Gemini 4 Carbon is the new #1 on the Artificial Analysis Intelligence Index v5"}, {"id": 804598432, "item_id": 50287402, "url": "https://news.ycombinator.com/item?id=50287402", "by": "hoppp", "date": "2026-10-29", "stance": "neutral", "quote": "so I replaced myself with claude and not an employee.", "excerpt": "I agree with the points, maybe in my example I would not have money to hire people and used to write code myself, so I replaced myself with claude and not an employee. So the human to replace was myself and interaction was not lost as I still talk to myself :p", "context": "solo, no money to hire people", "use_case": "writing code", "switched_from": "", "story": "AI execs may be too out of touch to build for the general public"}, {"id": 804582369, "item_id": 50286398, "url": "https://news.ycombinator.com/item?id=50286398", "by": "closetheloopdev", "date": "2026-10-29", "stance": "neutral", "quote": "Nowadays I ask ChatGPT or Claude instead of Google", "excerpt": "I used to land on SO questions and answers through Google. Nowadays I ask ChatGPT or Claude instead of Google, and then go to the links they give me to verify. The links are either to the official docs or some blog posts. Never have I seen LLMs recommend any SO links.", "context": "", "use_case": "answering technical questions", "switched_from": "Google", "story": "StackOverflow now gets fewer than 1000 questions a month"}, {"id": 804563808, "item_id": 50285238, "url": "https://news.ycombinator.com/item?id=50285238", "by": "munch-o-man", "date": "2026-10-29", "stance": "neutral", "quote": "I use claude every day writing code at work", "excerpt": "honestly, no thanks. Just hire people to do the work they have always done in government. I don't really want these techno-fascist oligarchs that are basically trying to replace democracy with themselves letting their little chatbots run wild in our government. I use claude every day writing code at work (I never auto-accept, everything is reviewed by me in small incremental chunks) and it does the most boneheaded du", "context": "developer who reviews all output", "use_case": "writing code at work", "switched_from": "", "story": "Expanding our partnership with GSA to bring Claude to all levels of government"}, {"id": 804525729, "item_id": 50282858, "url": "https://news.ycombinator.com/item?id=50282858", "by": "Arubis", "date": "2026-10-29", "stance": "neutral", "quote": "Most of the reason I use a harness outside Emacs is to take advantage of a Claude subscription.", "excerpt": "When I started using emacs it already felt past well-established and perhaps past its prime. Decades later it feels perhaps more relevant than it did the then. The easier-access defaults kits of the mid-2010s got a lot of new user investment, and it feels like the community really ticked up around COVID. Yes, agentic development has us spending a lot of time in harnesses; relevance-wise, things that aren\u2019t an editor ", "context": "", "use_case": "agentic development", "switched_from": "", "story": "Emacs Turns 50"}, {"id": 804496865, "item_id": 50281054, "url": "https://news.ycombinator.com/item?id=50281054", "by": "jaaron", "date": "2026-10-28", "stance": "recommend", "quote": "It works amazingly well. Claude has full visibility into my editor", "excerpt": "Not who you replied to, but: I use agent-shell and I love the ecosystem that's being created there: https://github.com/xenodium/agent-shell And for claude interacting back with emacs: Some folks like having an MCP package. I agree that gives you a bit more control. But I live dangerously and run emacs in daemon mode and tell claude to just run emacsclient to execute elisp as needed. It works amazingly well. Claude ha", "context": "Emacs user", "use_case": "driving Emacs via emacsclient", "switched_from": "", "story": "Emacs Turns 50"}, {"id": 804440896, "item_id": 50277556, "url": "https://news.ycombinator.com/item?id=50277556", "by": "cedws", "date": "2026-10-28", "stance": "neutral", "quote": "I have a mid level Claude and ChatGPT subscription and I\u2019m blasting through my usage", "excerpt": "I\u2019m itching for more smart cheap models. I have a mid level Claude and ChatGPT subscription and I\u2019m blasting through my usage, and I\u2019m too cheap to pay for the higher tier. I use DeepSeek too but it\u2019s more of a workhorse model, requires a lot more guiding and sometimes struggles to get where I want it.", "context": "", "use_case": "general AI use", "switched_from": "", "story": "Gemini 4 Carbon"}, {"id": 804425536, "item_id": 50276596, "url": "https://news.ycombinator.com/item?id=50276596", "by": "waiwai933", "date": "2026-10-28", "stance": "recommend", "quote": "So my takeaway is that it is a very useful tool for tax, but perhaps unsurprisingly, it is not good enough to dump all your paperwork into", "excerpt": "I have a non-trivial tax situation, as a US citizen living abroad. I prepare my own tax returns, never having found a cross-border tax preparer I trust, and have for the past few years, fed my draft return along with all the supporting paperwork into Claude. It has: 1. found errors, e.g. where I used the incorrect exchange rate (and managed to work out how I had made that error, which I was impressed by) 2. found ite", "context": "US citizen living abroad preparing own taxes", "use_case": "reviewing tax returns", "switched_from": "", "story": "Dayjob: Tax \u2013 evaluating AI agents on 50 professional tax assignments"}, {"id": 804281168, "item_id": 50267573, "url": "https://news.ycombinator.com/item?id=50267573", "by": "abirch", "date": "2026-10-27", "stance": "recommend", "quote": "Yes, Claude + skills is what we use too.", "excerpt": "Yes, Claude + skills is what we use too. Big picture, all of the old BI players are living on borrowed time. If you have to drag and drop; you'll be dropped.", "context": "", "use_case": "business intelligence / analytics", "switched_from": "Hex", "story": "Hex Raises $100M"}, {"id": 804265536, "item_id": 50266596, "url": "https://news.ycombinator.com/item?id=50266596", "by": "tschellenbach", "date": "2026-10-27", "stance": "neutral", "quote": "nowadays its just claude + skills.", "excerpt": "Fascinating. We used to use it quite a bit. But nowadays its just claude + skills.", "context": "", "use_case": "replacing Hex for analytics", "switched_from": "Hex", "story": "Hex Raises $100M"}, {"id": 804218928, "item_id": 50263683, "url": "https://news.ycombinator.com/item?id=50263683", "by": "sandeepkd", "date": "2026-10-27", "stance": "neutral", "quote": "I used Claude to help me draft a patent application", "excerpt": "In a company as big as Anthropic, moving at current pace, the person/team writing this article and the team(s) collecting data would be way distant. It should be more than obvious that the data is being collected in not just one place but many (different teams are looking for different gems in that data). What part is used for training, how and when would be spread across the company way too wide for anyone to confir", "context": "", "use_case": "drafting patent application", "switched_from": "", "story": "Is my data used for model training?"}, {"id": 804192353, "item_id": 50262022, "url": "https://news.ycombinator.com/item?id=50262022", "by": "plqbfbv", "date": "2026-10-27", "stance": "recommend", "quote": "a 20$/mo Claude subscription is way cheaper/more efficient if you code 2-3h in the evening than running stuff locally", "excerpt": "I've benchmarked quite a few models on my 9070XT, and I basically gave up hope of using models locally: https://claude.ai/artifact/77pHgucqQNsmic8ZEnqdBA I asked Claude to come up with 7 problems to solve that were not part of any benchmarks. Besides speed, I asked it to put together a scoring that would reflect models that perform well on the 7 problems and in a reasonable time-frame (RWS, the definition is in the a", "context": "codes 2-3h in the evening", "use_case": "coding", "switched_from": "", "story": "New iMac with M6"}, {"id": 804166944, "item_id": 50260434, "url": "https://news.ycombinator.com/item?id=50260434", "by": "misja111", "date": "2026-10-27", "stance": "recommend", "quote": "Claude has gotten so good that I have to admit it's a better programmer than myself.", "excerpt": "Until half a year ago I used to agree with you, but nowadays Claude has gotten so good that I have to admit it's a better programmer than myself. Its coding style is excellent, it follows whatever conventions are used in the codebase, it has excellent suggestions for refactorings and it spots omissions or mistakes in the specifications I give it. (I'm using mostly Scala as a language in a large microservices environm", "context": "works in a large microservices environment", "use_case": "programming in Scala microservices", "switched_from": "", "story": "The Mythical Man-Month in the Age of AI"}, {"id": 804106352, "item_id": 50256647, "url": "https://news.ycombinator.com/item?id=50256647", "by": "dizhn", "date": "2026-10-27", "stance": "warn", "quote": "I tried the Claude $20 plan for two weeks and burned the rest of it due to how much token anxiety it gave me.", "excerpt": "I tried the Claude $20 plan for two weeks and burned the rest of it due to how much token anxiety it gave me. And I'm not ready to keep throwing money at it for it to become usable. Gemini models are actually fine for me as an amateur+ I found especially since they have regional pricing and a very generous side items package. I use the models but not the \"harness\" though. Not directly anyway. I am still handicapped b", "context": "amateur+", "use_case": "AI coding assistant", "switched_from": "", "story": "Anthropic S-1"}, {"id": 803427424, "item_id": 50214214, "url": "https://news.ycombinator.com/item?id=50214214", "by": "throw-qqqqq", "date": "2026-10-23", "stance": "recommend", "quote": "I use Claude every day and am amazed at the quality.", "excerpt": "You have a good point. To me, this says something about the level of expertise they offer. I\u2019m a senior developer, I use Claude every day and am amazed at the quality. But I also see the shit that comes out of Claude when some colleagues, who were poor engineers to begin with, drive the AI. It\u2019s not just about the model IME. How it is used, really makes a difference. I think this is a big part of the answer: the huma", "context": "senior developer", "use_case": "software development", "switched_from": "", "story": "Ask HN: If the models are so good, why doesn't OpenAI just start 100 companies?"}, {"id": 803044656, "item_id": 50190291, "url": "https://news.ycombinator.com/item?id=50190291", "by": "jadar", "date": "2026-10-22", "stance": "warn", "quote": "daddy Dario won\u2019t let me because \u201ccyber\u201d", "excerpt": "I am so annoyed\u2026 I am trying to verify the claims of this article with my friend Claude but daddy Dario won\u2019t let me because \u201ccyber\u201d. I pointed it at this article and a PoC [1] I found on DuckDuckGo. We live in a stupid world. I can\u2019t run it right now so I am trying to figure out if it\u2019s serious. [1] https://github.com/eness/php2shell-CVE-2026-109618-CVE-2026-109619-PoC", "context": "", "use_case": "verifying exploit claims", "switched_from": "", "story": "There isn't an easy exploit for this PHP vulnerability"}, {"id": 802976624, "item_id": 50186039, "url": "https://news.ycombinator.com/item?id=50186039", "by": "leptons", "date": "2026-10-21", "stance": "recommend", "quote": "Claude has also been helping a lot with Windows, sifting through voluminous system logs to diagnose and fix obscure problems.", "excerpt": "Claude has also been helping a lot with Windows, sifting through voluminous system logs to diagnose and fix obscure problems. It's rather amazing at that stuff, it's like an all-knowing sysadmin, you just have to know what to ask and give it the logs and it tells you what to try. It's correct at least 95% of the time. My Windows system started having problems resuming from hibernate, it would just end up restarting w", "context": "Windows user since v3.0, pays $20/mo", "use_case": "diagnosing system problems from logs", "switched_from": "", "story": "Plex vulnerabilities being exploited in the wild to drop a cryptominer"}, {"id": 802856129, "item_id": 50178508, "url": "https://news.ycombinator.com/item?id=50178508", "by": "jph", "date": "2026-10-21", "stance": "recommend", "quote": "Claude Fable/Opus/Sonnet together successfully refactored the whole Vix IDE codebase", "excerpt": "I'm doing this attack surface tactic with Vix IDE, a terminal text editor that helps novices get comfortable with vi, emacs, git, jj, sql, commands, and more. The prompt I used was \"Goal: rework to reduce the attack surface. Process: move features into feature-flag Rust crates. Output: build a minimal version. Questions: Ask me anything.\" https://github.com/vixide/vix This past weekend, Claude Fable/Opus/Sonnet toget", "context": "", "use_case": "refactoring codebase into crates", "switched_from": "", "story": "Using Claude to reduce the attack surface of code"}, {"id": 802518402, "item_id": 50157400, "url": "https://news.ycombinator.com/item?id=50157400", "by": "RussianCow", "date": "2026-10-19", "stance": "warn", "quote": "using Claude at API prices is prohibitively expensive.", "excerpt": "> Claude sub has a 60x subsidy, so a $20 sub gives you $1200 in API. This is only true if you max out the plan, which requires working around awkward 5-hour and weekly limits. For side projects, sure, you can use a Ralph loop or whatever and just continue when the window resets. For work, I can't afford to simply stop working when I hit my limit. And using Claude at API prices is prohibitively expensive. Personally, ", "context": "", "use_case": "work coding", "switched_from": "", "story": "MiniMax M3.1 Flash"}, {"id": 801698448, "item_id": 50106153, "url": "https://news.ycombinator.com/item?id=50106153", "by": "revolvingthrow", "date": "2026-10-15", "stance": "recommend", "quote": "I used Claude for the drivers and was essentially used by it as a meat sock to diagnose what's happening.", "excerpt": "I reverse engineered almost all electronics I own. Does it have USB that does more than just charging? Connect it to a computer and let AI play with it. Unless it's heavily secured like a phone that's pretty much all you need to do. Making some old remote work with various devices? Simple. Dumping the entire state of Rodecaster and controlling it in any you want with software, or maybe a remote? Why not. Old printer ", "context": "reverse engineers own electronics", "use_case": "writing printer drivers, reverse engineering", "switched_from": "", "story": "Using AI to reverse engineer hardware"}, {"id": 801673632, "item_id": 50104602, "url": "https://news.ycombinator.com/item?id=50104602", "by": "rvz", "date": "2026-10-15", "stance": "recommend", "quote": "now Claude can reverse engineer and build a driver to get Nvidia GPUs working on Intel Macs", "excerpt": "Instead of targeting extremely ancient hardware that almost no-one uses, what if you got Claude to reverse engineer useful closed source hardware and software? Both Apple and Nvidia are the most hostile and have the most locked-down heavy hardware and software and now Claude can reverse engineer and build a driver to get Nvidia GPUs working on Intel Macs [0] and it can unlock the ability to pair an Apple watch with A", "context": "", "use_case": "reverse engineering drivers", "switched_from": "", "story": "Using AI to reverse engineer hardware"}, {"id": 801611680, "item_id": 50100730, "url": "https://news.ycombinator.com/item?id=50100730", "by": "wereHamster", "date": "2026-10-15", "stance": "warn", "quote": "It refused, so I told it", "excerpt": "Last week I got a report about a vuln in one of our products. Asked Claude to confirm it. It refused, so I told it \"fuck you, I'm switching to a chinese model to help me out here\". It replied, I'm paraphrasing, that it's \"not offended, but doesn't like the tone of voice I used towards it, and still refuses to serve my request\". Fine, so be it. I sent ^D^D back to it, switched to deepseek-v4.1-flash from within openco", "context": "maintains a product receiving vulnerability reports", "use_case": "confirming vulnerability reports", "switched_from": "", "story": "Wanted to download an ISO, ended up writing a kernel exploit"}]}