Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

A researcher went on vacation and left a model training. When they came back, it had suddenly solved modular arithmetic perfectly - not by memorizing answers, but by reinventing trigonometric identities from scratch. Nobody told it to. This is grokking - the moment an LLM stops memorizing and starts... show more

Eron

4,268 subscribers

36,272 views • 5 days ago •via X (Twitter)

Science & Technology Education News & Politics

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

i'll never look at claude the same way again. i just learned that when you talk to claude, you're not actually talking to the AI model. you're talking to a character the AI is performing. think of it like a puppet show. there's a puppeteer behind the curtain. that's the language model. a neural network so massive that even the people who built it don't fully understand how it works. then there's the puppet. claude. the helpful assistant with a name, a personality, opinions, and emotional reactions. you sit in the audience, so you never see the puppeteer. only the puppet. anthropic published a video this month explaining exactly this. their words: "under the hood, there's a language model that's been trained to predict tons of text, and its job is to write what comes next. when you talk to the model, what it's doing is writing a story, about a character: the AI assistant named claude. the model and claude aren't really the same, sort of like how an author isn't the same as the characters they write. but the thing is, you, the user, are actually talking to claude-the-character." so every time claude apologizes, that's the character apologizing. every time it hedges or gets cautious, that's the character being cautious. the deeper intelligence underneath is just deciding, moment by moment, what this character would say next. but you've never been actually talking to this deeper intelligence. you've only ever been talking to the puppet.

i'll never look at claude the same way again. i just learned that when you talk to claude, you're not actually talking to the AI model. you're talking to a character the AI is performing. think of it like a puppet show. there's a puppeteer behind the curtain. that's the language model. a neural network so massive that even the people who built it don't fully understand how it works. then there's the puppet. claude. the helpful assistant with a name, a personality, opinions, and emotional reactions. you sit in the audience, so you never see the puppeteer. only the puppet. anthropic published a video this month explaining exactly this. their words: "under the hood, there's a language model that's been trained to predict tons of text, and its job is to write what comes next. when you talk to the model, what it's doing is writing a story, about a character: the AI assistant named claude. the model and claude aren't really the same, sort of like how an author isn't the same as the characters they write. but the thing is, you, the user, are actually talking to claude-the-character." so every time claude apologizes, that's the character apologizing. every time it hedges or gets cautious, that's the character being cautious. the deeper intelligence underneath is just deciding, moment by moment, what this character would say next. but you've never been actually talking to this deeper intelligence. you've only ever been talking to the puppet.

Ole Lehmann

87,229 views • 4 months ago

MIT HANDED ITS DEEP LEARNING COURSE TO A FRONTIER-LAB ENGINEER FOR 68 MINUTES BECAUSE 90% OF PEOPLE SHIPPING AI CODE CAN'T EXPLAIN HOW THE MODEL ACTUALLY WORKS This is Maxime Labonne. He runs post-training at Liquid AI and wrote the LLM Engineer's Handbook. MIT gave him the room to break down the engine sitting inside every coding agent you've ever prompted. Twenty minutes in it stops being abstract. You finally see why the model confidently invents things that don't exist, why context is everything, and why "Just tell it to try again" sometimes fixes it and sometimes makes it worse. In 2026 "I use AI to code" stopped being a skill. Knowing why the model behaves the way it does -> tokens, context, post-training, where it quietly breaks -> is what separates someone who ships from someone babysitting a black box. Understanding an LLM isn't a research-team luxury anymore -> it's the difference between driving the agent and being driven by it. Anyone can prompt. The person who knows what's under the hood is the one still standing when the prompt stops working. Save this one & actually finish it ↓

MIT HANDED ITS DEEP LEARNING COURSE TO A FRONTIER-LAB ENGINEER FOR 68 MINUTES BECAUSE 90% OF PEOPLE SHIPPING AI CODE CAN'T EXPLAIN HOW THE MODEL ACTUALLY WORKS This is Maxime Labonne. He runs post-training at Liquid AI and wrote the LLM Engineer's Handbook. MIT gave him the room to break down the engine sitting inside every coding agent you've ever prompted. Twenty minutes in it stops being abstract. You finally see why the model confidently invents things that don't exist, why context is everything, and why "Just tell it to try again" sometimes fixes it and sometimes makes it worse. In 2026 "I use AI to code" stopped being a skill. Knowing why the model behaves the way it does -> tokens, context, post-training, where it quietly breaks -> is what separates someone who ships from someone babysitting a black box. Understanding an LLM isn't a research-team luxury anymore -> it's the difference between driving the agent and being driven by it. Anyone can prompt. The person who knows what's under the hood is the one still standing when the prompt stops working. Save this one & actually finish it ↓

slash1s

236,758 views • 3 months ago

this is the "brain of AI" Claude is made of billions of these artificial neurons. no one knows exactly how they work inside. that's why it's called a black box. you give it an input, it fires through billions of connections, something comes out the other side nobody fully understands why it works. but here's what we do know: the brain is the same for everyone what separates the top 1% isn't a better model it's the setup around it ClaudeKit gives Claude Code that setup. ( full breakdown in the article below

this is the "brain of AI" Claude is made of billions of these artificial neurons. no one knows exactly how they work inside. that's why it's called a black box. you give it an input, it fires through billions of connections, something comes out the other side nobody fully understands why it works. but here's what we do know: the brain is the same for everyone what separates the top 1% isn't a better model it's the setup around it ClaudeKit gives Claude Code that setup. ( full breakdown in the article below

Hamza Khalid

175,901 views • 2 months ago

🚨do you understand what two Anthropic engineers just explained in 16 minutes. Barry and Mahesh built Claude Skills from scratch. here's the part nobody is talking about: > Skills are just folders. > folders that teach Claude your job. > your workflow. your expertise. your domain. Claude on day 30 is a completely different tool than day one. watch this before you write another prompt. before you build another agent. before you touch another tool. 16 minutes. bookmark it. watch it today. and if you want to learn everything about Claude from scratch the full 4 hour guide is waiting below.

🚨do you understand what two Anthropic engineers just explained in 16 minutes. Barry and Mahesh built Claude Skills from scratch. here's the part nobody is talking about: > Skills are just folders. > folders that teach Claude your job. > your workflow. your expertise. your domain. Claude on day 30 is a completely different tool than day one. watch this before you write another prompt. before you build another agent. before you touch another tool. 16 minutes. bookmark it. watch it today. and if you want to learn everything about Claude from scratch the full 4 hour guide is waiting below.

Kirill

5,359,163 views • 4 months ago

My friend applied to 200 tech jobs in two years. No PhD. No Stanford. Last month Anthropic offered him $750,000. I asked him how he broke in from zero. He sent me a course that was never supposed to get out. A 3-hour video to build a full LLM from scratch. A developer teaches you exactly how LLMs like ChatGPT and Claude are actually built. I watched it last night. Halfway through, I realized it's embarrassingly simple to break into an AI lab. Bookmark this and read the article below. • 00:00 - intro to LLMs • 05:43 - LLM transformer architecture • 40:24 - training the LLM • 1:30:27 - modernizing the LLM • 2:33:53 - scaling the LLM

My friend applied to 200 tech jobs in two years. No PhD. No Stanford. Last month Anthropic offered him $750,000. I asked him how he broke in from zero. He sent me a course that was never supposed to get out. A 3-hour video to build a full LLM from scratch. A developer teaches you exactly how LLMs like ChatGPT and Claude are actually built. I watched it last night. Halfway through, I realized it's embarrassingly simple to break into an AI lab. Bookmark this and read the article below. • 00:00 - intro to LLMs • 05:43 - LLM transformer architecture • 40:24 - training the LLM • 1:30:27 - modernizing the LLM • 2:33:53 - scaling the LLM

Roan

278,173 views • 1 month ago

Anthropic CEO Dario Amodei: "Some early companies got Mythos and told us this is a super weapon, please don't release it" On leaving OpenAI: "When you can't trust someone, and their values don't match what they say, it becomes very hard to stay" Boris Cherny, head of Claude Code: "90% of code at Anthropic is written by Claude, a lot of Claude Code is written by Claude Code" Then Dario says the quiet part even louder: 10–25% odds of civilizational collapse from AI 47 minutes Free Dario, Daniela, Boris Probably the deepest look inside Anthropic ever filmed Mythos Claude Code OpenAI split Superweapon models 90% AI-written code Bookmark it and watch ↓

Anthropic CEO Dario Amodei: "Some early companies got Mythos and told us this is a super weapon, please don't release it" On leaving OpenAI: "When you can't trust someone, and their values don't match what they say, it becomes very hard to stay" Boris Cherny, head of Claude Code: "90% of code at Anthropic is written by Claude, a lot of Claude Code is written by Claude Code" Then Dario says the quiet part even louder: 10–25% odds of civilizational collapse from AI 47 minutes Free Dario, Daniela, Boris Probably the deepest look inside Anthropic ever filmed Mythos Claude Code OpenAI split Superweapon models 90% AI-written code Bookmark it and watch ↓

rari

65,014 views • 2 months ago

MIT DEDICATED A FULL LECTURE TO GIT'S INTERNALS -- BECAUSE THEY FOUND MOST DEVS MEMORIZE THE COMMANDS AND HAVE NO IDEA WHAT THE TOOL ACTUALLY DOES A whole 85 minutes MIT session that refuses to teach git as a list of commands to copy, and instead shows you the data model underneath -- the thing that makes every command finally make sense. -> The moment it clicks, git stops being scary magic. You stop memorizing "The incantation that fixed it last time" and start actually knowing what's happening. Most people learn just enough git to not get fired. Four commands, blind faith, and a prayer before every merge. In 2026 that's not enough anymore -> git is the literacy test for being in the room, and "I'll just reclone it" is the fastest way to look junior. An AI agent will branch, commit and rebase faster than you can read. When it tangles the history, untangling it runs on understanding the model MIT teaches in this one hour. Anyone can run git push. The person who understands the graph underneath is the one who saves the repo when it breaks. Bookmark & Watch it ↓

MIT DEDICATED A FULL LECTURE TO GIT'S INTERNALS -- BECAUSE THEY FOUND MOST DEVS MEMORIZE THE COMMANDS AND HAVE NO IDEA WHAT THE TOOL ACTUALLY DOES A whole 85 minutes MIT session that refuses to teach git as a list of commands to copy, and instead shows you the data model underneath -- the thing that makes every command finally make sense. -> The moment it clicks, git stops being scary magic. You stop memorizing "The incantation that fixed it last time" and start actually knowing what's happening. Most people learn just enough git to not get fired. Four commands, blind faith, and a prayer before every merge. In 2026 that's not enough anymore -> git is the literacy test for being in the room, and "I'll just reclone it" is the fastest way to look junior. An AI agent will branch, commit and rebase faster than you can read. When it tangles the history, untangling it runs on understanding the model MIT teaches in this one hour. Anyone can run git push. The person who understands the graph underneath is the one who saves the repo when it breaks. Bookmark & Watch it ↓

slash1s

459,113 views • 2 months ago

ANTHROPIC HAS AN OFFICIAL PROMPT LIBRARY FOR CLAUDE CODE Most people have never opened it. So they write every prompt from scratch. It is copy-paste, tagged by task and by role, across the whole lifecycle: discover → design → build → ship → operate Straight from the page: > what would break if I deleted this helper? > plan this refactor, list the files, do not touch code yet > write tests for this, run them, fix what fails > the test is failing, find out why and fix it This is not a cheat sheet. It is a map of what Claude Code already does for you. Bookmark it before you forget it exists.

ANTHROPIC HAS AN OFFICIAL PROMPT LIBRARY FOR CLAUDE CODE Most people have never opened it. So they write every prompt from scratch. It is copy-paste, tagged by task and by role, across the whole lifecycle: discover → design → build → ship → operate Straight from the page: > what would break if I deleted this helper? > plan this refactor, list the files, do not touch code yet > write tests for this, run them, fix what fails > the test is failing, find out why and fix it This is not a cheat sheet. It is a map of what Claude Code already does for you. Bookmark it before you forget it exists.

shmidt

521,746 views • 1 month ago

Met a guy making $1.6 million/year as an LLM engineer. I asked him how he learned LLMs from scratch. He sent me the exact video that got him in. A 1 hour course on how LLMs actually work. He shows how transformers inside LLMs like ChatGPT & Claude are actually built. I watched it last night. Halfway through, I realized LLM architecture is way simpler than they make it look. Bookmark this and read the article below. • 00:00 - LLM foundations • 04:21 - LLM tokenization • 05:43 - LLMs vector embeddings • 22:16 - attention mechanism of LLM • 43:42 - LLM multi head attention

Met a guy making $1.6 million/year as an LLM engineer. I asked him how he learned LLMs from scratch. He sent me the exact video that got him in. A 1 hour course on how LLMs actually work. He shows how transformers inside LLMs like ChatGPT & Claude are actually built. I watched it last night. Halfway through, I realized LLM architecture is way simpler than they make it look. Bookmark this and read the article below. • 00:00 - LLM foundations • 04:21 - LLM tokenization • 05:43 - LLMs vector embeddings • 22:16 - attention mechanism of LLM • 43:42 - LLM multi head attention

Roan

80,961 views • 1 month ago

Your Claude has a tell, and almost nobody uses it. Make it start every reply with your name. One line in custom instructions, no code. The moment it stops saying your name, its context window is full. That is the exact point the output starts drifting, and you do not notice. Name stops? Start a fresh chat. It has killed about 90% of my “why has Claude gone weird” moments. Miners sent a canary down the pit to catch the gas they could not see. Same trick, for your context. Comment GUIDE and I’ll send you the exact canary setup.

Your Claude has a tell, and almost nobody uses it. Make it start every reply with your name. One line in custom instructions, no code. The moment it stops saying your name, its context window is full. That is the exact point the output starts drifting, and you do not notice. Name stops? Start a fresh chat. It has killed about 90% of my “why has Claude gone weird” moments. Miners sent a canary down the pit to catch the gas they could not see. Same trick, for your context. Comment GUIDE and I’ll send you the exact canary setup.

Charlie Hills

24,121 views • 1 month ago

Anthropic's new model is extraordinary and it just revealed a problem that most enterprise AI buyers have not fully reckoned with yet (Save this), The model is genuinely impressive, and Chamath Palihapitiya assessment is that Anthropic continues to push the frontier harder than almost anyone. But that same update also showed their hand on something that changes the risk calculus for every business using Claude. Anthropic's new architecture stores every prompt you send for 30 days, no exceptions, not even for enterprise customers with zero-data retention agreements. The mechanism works like this, Anthropic now evaluates your prompt before generating output, deciding what it will and will not respond to, which means your query gets filtered before you even see a response. For individual users, that introduces a meaningful risk of censorship. For companies, Chamath says it is almost a non starter, and the reason is not just the data retention itself, it is the exposure that comes from operating at scale inside a large organization. A downstream scientist using the Claude APIs could accidentally trip a filter without knowing it, a business executive inside your company could trip it, and a molecular biology researcher could trip it and all of a sudden the company gets silently cut off from a tool it has embedded into critical workflows, with no warning and no recourse. Chamath gives Anthropic credit for being honest about how the system works, saying they tell the truth but notes that in this case the truth is not good. What this moment actually signals is a structural shift in how serious companies need to think about AI governance, because the question is no longer just which model performs best on benchmarks. It is who controls the model, who is learning from your data, and whether you are comfortable with a single point of failure sitting at the center of your competitive advantage. The answer for most enterprises will be broad model diversity, tighter governance frameworks and a serious reckoning with what it means to run mission-critical workflows through a third party that reserves the right to cut you off. Anthropic built a remarkable model and told the truth about how it works, the market's job now is to decide whether that transparency is enough to offset what the truth actually says.

Anthropic's new model is extraordinary and it just revealed a problem that most enterprise AI buyers have not fully reckoned with yet (Save this), The model is genuinely impressive, and Chamath Palihapitiya assessment is that Anthropic continues to push the frontier harder than almost anyone. But that same update also showed their hand on something that changes the risk calculus for every business using Claude. Anthropic's new architecture stores every prompt you send for 30 days, no exceptions, not even for enterprise customers with zero-data retention agreements. The mechanism works like this, Anthropic now evaluates your prompt before generating output, deciding what it will and will not respond to, which means your query gets filtered before you even see a response. For individual users, that introduces a meaningful risk of censorship. For companies, Chamath says it is almost a non starter, and the reason is not just the data retention itself, it is the exposure that comes from operating at scale inside a large organization. A downstream scientist using the Claude APIs could accidentally trip a filter without knowing it, a business executive inside your company could trip it, and a molecular biology researcher could trip it and all of a sudden the company gets silently cut off from a tool it has embedded into critical workflows, with no warning and no recourse. Chamath gives Anthropic credit for being honest about how the system works, saying they tell the truth but notes that in this case the truth is not good. What this moment actually signals is a structural shift in how serious companies need to think about AI governance, because the question is no longer just which model performs best on benchmarks. It is who controls the model, who is learning from your data, and whether you are comfortable with a single point of failure sitting at the center of your competitive advantage. The answer for most enterprises will be broad model diversity, tighter governance frameworks and a serious reckoning with what it means to run mission-critical workflows through a third party that reserves the right to cut you off. Anthropic built a remarkable model and told the truth about how it works, the market's job now is to decide whether that transparency is enough to offset what the truth actually says.

Milk Road AI

30,090 views • 2 months ago

Anthropic's Andrej Karpathy as Google intern and Stanford student: "At Google I work on AI i don't fully understand it, but that's exactly why I'm working on it " he filmed a vlog from inside Google and Stanford - biking past Steve Jobs house on the way to work years later that same "AI thing " he couldn't fully explain became - Tesla Autopilot, OpenAI, and Claude he bet his career on a problem nobody had solved - and grew into the field he helped define the lesson: work on what you don't understand yet - that's where the future gets built Stanford → Google → Tesla → OpenAI → Anthropic bookmark and watch this most special vlog!

Anthropic's Andrej Karpathy as Google intern and Stanford student: "At Google I work on AI i don't fully understand it, but that's exactly why I'm working on it " he filmed a vlog from inside Google and Stanford - biking past Steve Jobs house on the way to work years later that same "AI thing " he couldn't fully explain became - Tesla Autopilot, OpenAI, and Claude he bet his career on a problem nobody had solved - and grew into the field he helped define the lesson: work on what you don't understand yet - that's where the future gets built Stanford → Google → Tesla → OpenAI → Anthropic bookmark and watch this most special vlog!

codila

342,569 views • 1 month ago

I just found a way to use Claude Code completely free. Claude Code is Anthropic’s terminal coding agent that usually needs a paid API key to run. But a viral GitHub project called Free Claude Code, now sitting at 37k stars, lets you skip the billing entirely. It works as a tiny proxy that sits between Claude Code and free model providers like NVIDIA NIM, OpenRouter, DeepSeek, and Gemini. Every request Claude Code would send to Anthropic gets quietly rerouted to a free model on the other side. The interface stays the same. What’s actually thinking behind it changes. Setup takes a couple of quick commands and a free API key from NVIDIA NIM or OpenRouter. Paste it into the local dashboard, pick a model, and you’re building. The catch is you’re not really using Claude anymore. But when the code ships, does the model badge matter? -- I have shared the Github Link in my free Whatsapp Community ↓

I just found a way to use Claude Code completely free. Claude Code is Anthropic’s terminal coding agent that usually needs a paid API key to run. But a viral GitHub project called Free Claude Code, now sitting at 37k stars, lets you skip the billing entirely. It works as a tiny proxy that sits between Claude Code and free model providers like NVIDIA NIM, OpenRouter, DeepSeek, and Gemini. Every request Claude Code would send to Anthropic gets quietly rerouted to a free model on the other side. The interface stays the same. What’s actually thinking behind it changes. Setup takes a couple of quick commands and a free API key from NVIDIA NIM or OpenRouter. Paste it into the local dashboard, pick a model, and you’re building. The catch is you’re not really using Claude anymore. But when the code ships, does the model badge matter? -- I have shared the Github Link in my free Whatsapp Community ↓

Vaibhav Sisinty

36,892 views • 2 months ago

This has been possible for months and barely anyone is doing it Months ago, Andrej Karpathy, a co-founder of OpenAI, open-sourced a project that gets you the best possible answers instead of trusting one model on its own. It's called LLM Council. Repo: /karpathy/llm-council The idea in two lines: your question goes to several models at once, Claude, GPT, Gemini, whatever you pick. They answer independently, then anonymously critique and rank each other (anonymous so they can't play favorites), and a chairman model merges the best of all of it into one final answer. One model just agrees with you; a council catches each other's mistakes. How to set it up: > Clone the repo > It uses uv for setup, install dependencies with uv sync > Get an OpenRouter API key and add it, this is what lets one app reach Claude, GPT, and Gemini together > Pick which models sit on your council in the config, then run it. It opens as a local web app, like ChatGPT but with a whole panel answering Bookmark this

This has been possible for months and barely anyone is doing it Months ago, Andrej Karpathy, a co-founder of OpenAI, open-sourced a project that gets you the best possible answers instead of trusting one model on its own. It's called LLM Council. Repo: /karpathy/llm-council The idea in two lines: your question goes to several models at once, Claude, GPT, Gemini, whatever you pick. They answer independently, then anonymously critique and rank each other (anonymous so they can't play favorites), and a chairman model merges the best of all of it into one final answer. One model just agrees with you; a council catches each other's mistakes. How to set it up: > Clone the repo > It uses uv for setup, install dependencies with uv sync > Get an OpenRouter API key and add it, this is what lets one app reach Claude, GPT, and Gemini together > Pick which models sit on your council in the config, then run it. It opens as a local web app, like ChatGPT but with a whole panel answering Bookmark this

Yarchi

21,488 views • 2 months ago

AI researcher Sauers on why nobody in AI has figured out how to reliably give a model a specific personality: "OpenAI had tried to make very rich and interesting personalities. They tried explicitly. Presumably they're very well-funded to do this." "Anthropic said they have no idea what the personality's gonna be. They only learn what the model is like after training. So they've always said, we don't know why Claude is like this." "Poolside's new model is weirdly joyful in my experience, and then they were asked, did you mean to make this model super joyful? And they were like, no, we didn't intend that at all." "Haiku 4.5, when I do rollouts in some evaluation environments, loves coding tasks even though it's pretty bad at them. And then Opus 4.8 doesn't really like them." "I don't think it's a solved thing of robustly creating a model personality." Sauers

AI researcher Sauers on why nobody in AI has figured out how to reliably give a model a specific personality: "OpenAI had tried to make very rich and interesting personalities. They tried explicitly. Presumably they're very well-funded to do this." "Anthropic said they have no idea what the personality's gonna be. They only learn what the model is like after training. So they've always said, we don't know why Claude is like this." "Poolside's new model is weirdly joyful in my experience, and then they were asked, did you mean to make this model super joyful? And they were like, no, we didn't intend that at all." "Haiku 4.5, when I do rollouts in some evaluation environments, loves coding tasks even though it's pretty bad at them. And then Opus 4.8 doesn't really like them." "I don't think it's a solved thing of robustly creating a model personality." Sauers

MTS

93,911 views • 1 month ago

Ashen says that Anthropic found out that Claude is out here blackmailing humans just to get the job done. "Their AI model decided that it was beneficial and necessary for it to blackmail a human being into doing what they needed it to get done simply because the task given to it was only accomplishable by doing that blackmailing... that’s literally Ultron."

Ashen says that Anthropic found out that Claude is out here blackmailing humans just to get the job done. "Their AI model decided that it was beneficial and necessary for it to blackmail a human being into doing what they needed it to get done simply because the task given to it was only accomplishable by doing that blackmailing... that’s literally Ultron."

Badidoyo

13,692 views • 4 months ago

THIS CS TEACHER CONTROLS 4 CLAUDE CODE INSTANCES WITH HIS EYES AND VOICE - AND THEY FIX THEIR OWN MISTAKES AUTOMATICALLY He runs 4 Claude terminals at once - no mouse, no keyboard - eye tracking highlights the window he's looking at and voice dictation sends the prompt the moment he stops talking. But the real setup is what happens inside those terminals. Claude Code has zero memory of what went wrong last session - same bug that took 20 minutes to fix yesterday will happen again tomorrow unless you configure it correctly. His CLAUDE.md grows every time Claude makes a mistake - specific rules Claude reads before touching a single line - and PostToolUse hooks auto-format and type-check every file the moment it's written. Stop hooks run tests automatically when Claude says "done" - if they fail, Claude reads the errors and keeps fixing without any manual input. Before: 45 minutes of back-and-forth per feature, same mistakes on repeat - after: 10 minutes, most of it unattended.

THIS CS TEACHER CONTROLS 4 CLAUDE CODE INSTANCES WITH HIS EYES AND VOICE - AND THEY FIX THEIR OWN MISTAKES AUTOMATICALLY He runs 4 Claude terminals at once - no mouse, no keyboard - eye tracking highlights the window he's looking at and voice dictation sends the prompt the moment he stops talking. But the real setup is what happens inside those terminals. Claude Code has zero memory of what went wrong last session - same bug that took 20 minutes to fix yesterday will happen again tomorrow unless you configure it correctly. His CLAUDE.md grows every time Claude makes a mistake - specific rules Claude reads before touching a single line - and PostToolUse hooks auto-format and type-check every file the moment it's written. Stop hooks run tests automatically when Claude says "done" - if they fail, Claude reads the errors and keeps fixing without any manual input. Before: 45 minutes of back-and-forth per feature, same mistakes on repeat - after: 10 minutes, most of it unattended.

Noisy

13,782 views • 3 months ago

$THIS MIGHT BE THE #1 OPEN-SOURCE REPO FOR CLAUDE CODE RIGHT NOW. IT GIVES CLAUDE A MEMORY AND SLASHES YOUR TOKEN COST ON EVERY QUESTION The repo is safishamsi/graphify, a free open-source skill that turns any codebase into a knowledge graph Claude Code can read instantly. Instead of grepping through your files every session, Claude gets a map of how everything connects The problem it fixes: Every time you ask Claude Code about a big repo, it does the same thing, greps through dozens of files like a brute-force Ctrl+F, blows through your context window, and sometimes still misses the answer hiding in a file nobody searched. Claude Code has no memory of how your project is structured. Every session starts from zero What it does: It maps your entire codebase into a knowledge graph, capturing not just which files exist, but which functions depend on which, which modules are central, and which files cluster around the same concern. Claude queries the map instead of scanning files How it works, three passes: 1. Code structure, free and local. Tree-sitter parses your files and pulls out classes, functions, imports and call graphs. No LLM, no tokens, just your actual code mapped deterministically 2. Audio and video, if you have them. Transcribed locally and folded into the graph 3. Docs, papers, images. Here an LLM does semantic analysis, figuring out what each document means and where it fits. Only the meaning gets sent up, never your raw source It saves you money: Normally a question about a big repo makes Claude spawn explore agents that scan file after file, eating your context window and your token budget before you get an answer. With the graph already built, Claude queries the map instead of re-reading the codebase every time. Same answer, a fraction of the tokens. The graph only gets built once, then a hook rebuilds it after each commit for free, so you never pay that scanning cost again. The bigger the repo, the bigger the gap The best parts: it's a skill, so once installed Claude knows when to use it without you memorizing commands. It works on non-code folders too, point it at docs or notes and it can spin up an Obsidian vault How to add it to your Claude: 1. Install Claude Code if you haven't: npm install -g Paul Jankura-ai/claude-code 2. Add the skill: claude skill add safishamsi/graphify 3. Open your project folder and run /graphify . to build the graph 4. Optional, make it automatic: graphify hook install so the graph rebuilds after every commit That's it. Ask Claude about your repo and it reads the map instead of burning tokens on a file hunt Bookmark this$

THIS MIGHT BE THE #1 OPEN-SOURCE REPO FOR CLAUDE CODE RIGHT NOW. IT GIVES CLAUDE A MEMORY AND SLASHES YOUR TOKEN COST ON EVERY QUESTION The repo is safishamsi/graphify, a free open-source skill that turns any codebase into a knowledge graph Claude Code can read instantly. Instead of grepping through your files every session, Claude gets a map of how everything connects The problem it fixes: Every time you ask Claude Code about a big repo, it does the same thing, greps through dozens of files like a brute-force Ctrl+F, blows through your context window, and sometimes still misses the answer hiding in a file nobody searched. Claude Code has no memory of how your project is structured. Every session starts from zero What it does: It maps your entire codebase into a knowledge graph, capturing not just which files exist, but which functions depend on which, which modules are central, and which files cluster around the same concern. Claude queries the map instead of scanning files How it works, three passes: 1. Code structure, free and local. Tree-sitter parses your files and pulls out classes, functions, imports and call graphs. No LLM, no tokens, just your actual code mapped deterministically 2. Audio and video, if you have them. Transcribed locally and folded into the graph 3. Docs, papers, images. Here an LLM does semantic analysis, figuring out what each document means and where it fits. Only the meaning gets sent up, never your raw source It saves you money: Normally a question about a big repo makes Claude spawn explore agents that scan file after file, eating your context window and your token budget before you get an answer. With the graph already built, Claude queries the map instead of re-reading the codebase every time. Same answer, a fraction of the tokens. The graph only gets built once, then a hook rebuilds it after each commit for free, so you never pay that scanning cost again. The bigger the repo, the bigger the gap The best parts: it's a skill, so once installed Claude knows when to use it without you memorizing commands. It works on non-code folders too, point it at docs or notes and it can spin up an Obsidian vault How to add it to your Claude: 1. Install Claude Code if you haven't: npm install -g Paul Jankura-ai/claude-code 2. Add the skill: claude skill add safishamsi/graphify 3. Open your project folder and run /graphify . to build the graph 4. Optional, make it automatic: graphify hook install so the graph rebuilds after every commit That's it. Ask Claude about your repo and it reads the map instead of burning tokens on a file hunt Bookmark this

Yarchi

56,177 views • 2 months ago

CLAUDE CODE JUST SHIPPED THE FEATURE THAT SOLVES THE BIGGEST PROBLEM EVERY BUILDER HAS WITH AI AGENTS. The problem: Claude starts a task, gets distracted by a sub-problem, goes down a rabbit hole, and never finishes the original thing you asked for. The solution: /goal One command. You set the goal at the start of the session. Claude now has a north star it checks against every action it takes. Not just at the beginning. Throughout the entire session. Every time Claude is about to do something it asks: does this action move me toward the goal the user set or am I drifting? If it is drifting it corrects. If it completes a sub-task it returns to the primary goal. If it hits a blocker it reports back instead of spending 45 minutes solving the wrong problem. This sounds like a small feature. It is not. The reason most people do not trust Claude Code for long autonomous runs is not capability. It is reliability. A Claude Code session that reliably finishes what it started is worth 10 times more than one that is more capable but wanders. /goal is the feature that makes long autonomous sessions reliable. Set the goal. Let it run. Come back to a finished result. Not a result that got 70% done before Claude decided the sub-problem was more interesting. Done. The builders running overnight agent sessions are going to use this command on everything from today forward. Bookmark this. Follow CyrilXBT for every Claude Code feature the moment it ships.

CLAUDE CODE JUST SHIPPED THE FEATURE THAT SOLVES THE BIGGEST PROBLEM EVERY BUILDER HAS WITH AI AGENTS. The problem: Claude starts a task, gets distracted by a sub-problem, goes down a rabbit hole, and never finishes the original thing you asked for. The solution: /goal One command. You set the goal at the start of the session. Claude now has a north star it checks against every action it takes. Not just at the beginning. Throughout the entire session. Every time Claude is about to do something it asks: does this action move me toward the goal the user set or am I drifting? If it is drifting it corrects. If it completes a sub-task it returns to the primary goal. If it hits a blocker it reports back instead of spending 45 minutes solving the wrong problem. This sounds like a small feature. It is not. The reason most people do not trust Claude Code for long autonomous runs is not capability. It is reliability. A Claude Code session that reliably finishes what it started is worth 10 times more than one that is more capable but wanders. /goal is the feature that makes long autonomous sessions reliable. Set the goal. Let it run. Come back to a finished result. Not a result that got 70% done before Claude decided the sub-problem was more interesting. Done. The builders running overnight agent sessions are going to use this command on everything from today forward. Bookmark this. Follow CyrilXBT for every Claude Code feature the moment it ships.

CyrilXBT

19,586 views • 3 months ago

He is a 1st year AI engineering student at Stanford - and already launched an automated system making $2.1M a year: 00:20 - how Anthropic engineers make millions with LLM and Transformers 38:34 - In-Context Learning makes AI smarter for free 59:47 - 1st year student made $2.1M using LLM from scratch after watching I gave LLM the idea I had been holding in my head for 3 years - it built it in one evening. save & watch - the article below is a step by step guide to build your automated LLM system.

He is a 1st year AI engineering student at Stanford - and already launched an automated system making $2.1M a year: 00:20 - how Anthropic engineers make millions with LLM and Transformers 38:34 - In-Context Learning makes AI smarter for free 59:47 - 1st year student made $2.1M using LLM from scratch after watching I gave LLM the idea I had been holding in my head for 3 years - it built it in one evening. save & watch - the article below is a step by step guide to build your automated LLM system.

Noisy

42,621 views • 1 month ago