Загрузка видео...

Не удалось загрузить видео

На главную

MLX + TurboQuant = Local Super Power Take a local (private) document(s) or codebase, pre-fill the 256k KV cache (the context) with the document(s) and system prompt, quantise and run on Apple's MLX and you have almost instantaneous, lossless document queries with total privacy. For a 75-page PDF (some...

jtdavies's profile picture

John T Davies 🇪🇺

3,211 subscribers

85,778 просмотров • 5 месяцев назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

Introducing Magnitude: your actually local agent 100% private and offline. No token costs, no API keys. Open source. Today's agents are local. The model isn't. Every prompt, every file, every secret gets sent straight to Anthropic and OpenAI. Magnitude is built around local models and runs the whole stack itself. The inference engine is part of the agent, so the models run inside it, right on your computer. It lives in your terminal, and setup is one command. Magnitude profiles your hardware and shows you which models fit, with the trade-offs between quality, speed, and memory. Pick one and start working. No painful config or server to babysit. Out of the box, it can use your shell, edit files, and run scripts. Add skills and it can work with Excel, PowerPoint, PDFs or Chrome. Use it for everyday work: - Analyze sensitive data - Manage private notes - Review code and logs - Search and organize files - Build docs or slides npm i -g @magnitudedev/cli GitHub:
0:35

Introducing Magnitude: your actually local agent 100% private and offline. No token costs, no API keys. Open source. Today's agents are local. The model isn't. Every prompt, every file, every secret gets sent straight to Anthropic and OpenAI. Magnitude is built around local models and runs the whole stack itself. The inference engine is part of the agent, so the models run inside it, right on your computer. It lives in your terminal, and setup is one command. Magnitude profiles your hardware and shows you which models fit, with the trade-offs between quality, speed, and memory. Pick one and start working. No painful config or server to babysit. Out of the box, it can use your shell, edit files, and run scripts. Add skills and it can work with Excel, PowerPoint, PDFs or Chrome. Use it for everyday work: - Analyze sensitive data - Manage private notes - Review code and logs - Search and organize files - Build docs or slides npm i -g @magnitudedev/cli GitHub:

Tom Greenwald

397,093 просмотров • 1 месяц назад

A small project I'm publishing for Neovim people: 🐠 butterfish.nvim This is a Neovim plugin to use language models for very fluent code editing. It veers away from the chatbox or codebase Q/A approaches and instead focuses on specific local edits, for example, select and rewrite a block of code. Do powerful things with few keypresses, without leaving the Neovim buffer!
4:56

A small project I'm publishing for Neovim people: 🐠 butterfish.nvim This is a Neovim plugin to use language models for very fluent code editing. It veers away from the chatbox or codebase Q/A approaches and instead focuses on specific local edits, for example, select and rewrite a block of code. Do powerful things with few keypresses, without leaving the Neovim buffer!

Peter Bakkum

29,074 просмотров • 2 лет назад

$200,000 a year on a dev team. Fable 5 does the same thing in one night for $0. My friend who works as a Claude Code engineer made $700,000 without a single developer on his team. 10:27 One person wrote one PRD document. Fable 5 did the rest - 21,000 lines of TypeScript, 90+ commits, a full 3D game in the browser with zero downloads. Anthropic gave free access to this for 7 days and nobody is talking about it. I watched the demo and couldn't sleep. For two years I was hiring people for what one document and one night can do better. Watch the video below then read the article.
11:58

$200,000 a year on a dev team. Fable 5 does the same thing in one night for $0. My friend who works as a Claude Code engineer made $700,000 without a single developer on his team. 10:27 One person wrote one PRD document. Fable 5 did the rest - 21,000 lines of TypeScript, 90+ commits, a full 3D game in the browser with zero downloads. Anthropic gave free access to this for 7 days and nobody is talking about it. I watched the demo and couldn't sleep. For two years I was hiring people for what one document and one night can do better. Watch the video below then read the article.

Noisy

11,153 просмотров • 2 месяцев назад

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
13:23

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

56,502 просмотров • 3 месяцев назад

President of Ukraine: We did indeed have a positive meeting with President Trump and his team. First of all, to be honest, I raised the issue of air defense. And I think that the US will return with a positive result. I will definitely talk about it when the result is in Ukraine. Second, our security guarantees. They are indeed ready, and the agreement is ready for signing. Of course, there will be many different technical details. And based on this document, additional documents will appear, but the main, core agreement on security guarantees is in place. Now I am waiting for President Trump to set the date and place. It's up to him. We are ready to sign such an important [document] for us. I emphasize once again that I am confident this is a historic document. Why did I mention the many details? You know that such a document awaits ratification by the US Congress and the Ukrainian Parliament. Therefore, there are still many steps ahead on this path.
1:09

President of Ukraine: We did indeed have a positive meeting with President Trump and his team. First of all, to be honest, I raised the issue of air defense. And I think that the US will return with a positive result. I will definitely talk about it when the result is in Ukraine. Second, our security guarantees. They are indeed ready, and the agreement is ready for signing. Of course, there will be many different technical details. And based on this document, additional documents will appear, but the main, core agreement on security guarantees is in place. Now I am waiting for President Trump to set the date and place. It's up to him. We are ready to sign such an important [document] for us. I emphasize once again that I am confident this is a historic document. Why did I mention the many details? You know that such a document awaits ratification by the US Congress and the Ukrainian Parliament. Therefore, there are still many steps ahead on this path.

Anton Gerashchenko

27,842 просмотров • 8 месяцев назад

The same kinds of productivity gains we've seen in coding with AI agents are heading to the rest of knowledge work. This is the jump when you go from having a chatbot to being able to actually have an agent go off and do work for minutes or even hours and come back with a complete work output that you then review. Here's an example of the new Box Agent filling out an RFP response from an existing knowledge base. This process would normally take hours to fill out, and requires the full attention of the user doing the work. Now, you provide the Box Agent with the RFP questions, and it will go off, make a plan, extract all the relevant questions, read through existing source material to come up with an answer, and then generate a new word document as the final output. All while you're doing something else. The key to this architecture is that the agent is able to use all of the same tools in the background that a user uses to get work done. The agent can search for documents, read entire files, run scripts and tools in the background, and even be able to write code on the fly to automate tasks it hasn't seen before. And best of all, the Box Agent will (soon) work from the Box MCP and CLI so you can invoke it in any agentic system as a step in a process. This kind of agent complexity would have been impossible even 6 months ago. Models consistently failed at tracking long running tasks or using the right tools at the right moment for the task. But this is all now possible because of models like GPT-5.4, Opus 4.6, and Gemini 3, and is only getting better by the month. Just as we moved from engineers writing code and using AI as an assistant to answer questions, in many areas of knowledge work -like legal, finance, consulting, sales, marketing, and more- when we have a problem we'll just kick off the AI agent to just go work on it for us in the background.
2:14

The same kinds of productivity gains we've seen in coding with AI agents are heading to the rest of knowledge work. This is the jump when you go from having a chatbot to being able to actually have an agent go off and do work for minutes or even hours and come back with a complete work output that you then review. Here's an example of the new Box Agent filling out an RFP response from an existing knowledge base. This process would normally take hours to fill out, and requires the full attention of the user doing the work. Now, you provide the Box Agent with the RFP questions, and it will go off, make a plan, extract all the relevant questions, read through existing source material to come up with an answer, and then generate a new word document as the final output. All while you're doing something else. The key to this architecture is that the agent is able to use all of the same tools in the background that a user uses to get work done. The agent can search for documents, read entire files, run scripts and tools in the background, and even be able to write code on the fly to automate tasks it hasn't seen before. And best of all, the Box Agent will (soon) work from the Box MCP and CLI so you can invoke it in any agentic system as a step in a process. This kind of agent complexity would have been impossible even 6 months ago. Models consistently failed at tracking long running tasks or using the right tools at the right moment for the task. But this is all now possible because of models like GPT-5.4, Opus 4.6, and Gemini 3, and is only getting better by the month. Just as we moved from engineers writing code and using AI as an assistant to answer questions, in many areas of knowledge work -like legal, finance, consulting, sales, marketing, and more- when we have a problem we'll just kick off the AI agent to just go work on it for us in the background.

Aaron Levie

24,728 просмотров • 5 месяцев назад

The 20 points for ending the war form a fundamental document. We are actively working on the key steps – they must be doable. From this fundamental document, we are developing at least two additional ones. The first is on security – regarding security guarantees with the United States. The second is on the economy, covering reconstruction and joint investment. Today, we focused specifically on the economic document. The principles of the economic document are absolutely clear, and we are fully aligned with the American side on this. Thank you. An economic action plan will follow. Europe will also be involved in reconstruction. In the near future, we plan to move forward on the other two documents as well. Rustem Umerov will continue the discussions.
4:34

The 20 points for ending the war form a fundamental document. We are actively working on the key steps – they must be doable. From this fundamental document, we are developing at least two additional ones. The first is on security – regarding security guarantees with the United States. The second is on the economy, covering reconstruction and joint investment. Today, we focused specifically on the economic document. The principles of the economic document are absolutely clear, and we are fully aligned with the American side on this. Thank you. An economic action plan will follow. Europe will also be involved in reconstruction. In the near future, we plan to move forward on the other two documents as well. Rustem Umerov will continue the discussions.

Volodymyr Zelenskyy / Володимир Зеленський

470,384 просмотров • 9 месяцев назад

Claude Tag has completely changed the way I do work for the last 4 months. Except… it's not Claude Tag. Anthropic only announced that a few hours ago, and I don't even have access yet. But I did build a version of it for myself which I've been using for months now. Here's how. 4 months ago, inspired by the success of OpenClaw, I wondered what would happen if I let Claude Code on its own computer 24x7. So I built a simple harness that allowed me to turn any Mac into an AI employee with Claude Code headless mode (-p). Today, I manage 3 such AI employees. It started with Luo Ji — my and my brother Piyush Agarwal's AI co-founder, running in our personal Slack. Luo does real work for us. We've been writing a 100% of the code for 3 products on Slack with Luo now. It manages our emails and gives us a little brief each day with things we need to take action on. And so much more. And it's not just the two of us. On the consulting team at Every 🪨, we run Claudie and for the editorial team, Andy. Same architecture, same Slack, months of real work. They help the teams with work related to project management, chief-of-staff work, data hygiene, building decks, writing first drafts, even browsing X on their own account for AI updates. So it's mindblowing to see that Anthropic landed on the exact same architecture I did. Claude Tag is an AI employee that lives in your Slack workspace and does work autonomously. Anthropic says they've been running it internally for the better part of this year — opening PRs, doing real work. And so have I. So has my whole team. The architectural decisions Anthropic baked into Claude Tag are the ones we arrived at too: - Built on Claude Code - Uses its own accounts - A separate employee per team - Slack as the interface This is the future of work, and I've been living it for months. I've shifted all of my workflows — code, PRs, even the non-technical stuff — out of Claude Code and the Claude app and into Slack. I've had entire weeks where I never opened Claude Code on my laptop. Here's a video walkthrough of how I've been using this in real life.
14:36

Claude Tag has completely changed the way I do work for the last 4 months. Except… it's not Claude Tag. Anthropic only announced that a few hours ago, and I don't even have access yet. But I did build a version of it for myself which I've been using for months now. Here's how. 4 months ago, inspired by the success of OpenClaw, I wondered what would happen if I let Claude Code on its own computer 24x7. So I built a simple harness that allowed me to turn any Mac into an AI employee with Claude Code headless mode (-p). Today, I manage 3 such AI employees. It started with Luo Ji — my and my brother Piyush Agarwal's AI co-founder, running in our personal Slack. Luo does real work for us. We've been writing a 100% of the code for 3 products on Slack with Luo now. It manages our emails and gives us a little brief each day with things we need to take action on. And so much more. And it's not just the two of us. On the consulting team at Every 🪨, we run Claudie and for the editorial team, Andy. Same architecture, same Slack, months of real work. They help the teams with work related to project management, chief-of-staff work, data hygiene, building decks, writing first drafts, even browsing X on their own account for AI updates. So it's mindblowing to see that Anthropic landed on the exact same architecture I did. Claude Tag is an AI employee that lives in your Slack workspace and does work autonomously. Anthropic says they've been running it internally for the better part of this year — opening PRs, doing real work. And so have I. So has my whole team. The architectural decisions Anthropic baked into Claude Tag are the ones we arrived at too: - Built on Claude Code - Uses its own accounts - A separate employee per team - Slack as the interface This is the future of work, and I've been living it for months. I've shifted all of my workflows — code, PRs, even the non-technical stuff — out of Claude Code and the Claude app and into Slack. I've had entire weeks where I never opened Claude Code on my laptop. Here's a video walkthrough of how I've been using this in real life.

Nityesh

36,790 просмотров • 3 месяцев назад

This Chinese developer runs 9 agents on Claude Code under a GPT-5.5 orchestrator and they close 500 client tasks a month without a single assistant. His client work is closed without him, on a single laptop and only three subscriptions. The entire system lives on one MacBook Pro M4 with 128 GB of memory and subscriptions to Claude Code and GPT-5.5 cost him approximately $300 a month. There is no CRM, no team, no office only a terminal window with 9 parallel streams. The orchestrator works with a simple system prompt: «You are the orchestrator of a client inbox. Classify every incoming email into 4 categories: code, content, analysis, communication. Delegate to the corresponding worker agent. When the result is ready, check it for completeness, send it to the client on my behalf, and mark the task as closed. Do not ask clarifying questions.» And the orchestrator checks the inbox every 30 seconds, classifies fresh emails, and distributes them to 9 worker agents on Claude Code, each of whom is responsible for their own class of tasks. Here is an example of how one of them closes a request to refactor a client's auth module: Task: refactor user-auth module Broke the monolith into 3 files by responsibilities Added unit tests, coverage increased to 87% Renamed 4 functions to camelCase according to the style guide PR is ready for review, link below» And so about 50 cycles a day. By noon 25 tasks are closed, by dinner 50, and by the end of the month 500. On average, it takes about 7 minutes from the appearance of an email in the inbox to sending the result to the client. This is more than what a live team of 6 developers, copywriters and analysts working 8 hours a day closes. This is no longer an agency. This is a workstation where an orchestrator replaces a manager, and 9 worker agents replace the staff. The pipeline goes from inbox to closing 500 times a month without human participation at any step.
1:32

This Chinese developer runs 9 agents on Claude Code under a GPT-5.5 orchestrator and they close 500 client tasks a month without a single assistant. His client work is closed without him, on a single laptop and only three subscriptions. The entire system lives on one MacBook Pro M4 with 128 GB of memory and subscriptions to Claude Code and GPT-5.5 cost him approximately $300 a month. There is no CRM, no team, no office only a terminal window with 9 parallel streams. The orchestrator works with a simple system prompt: «You are the orchestrator of a client inbox. Classify every incoming email into 4 categories: code, content, analysis, communication. Delegate to the corresponding worker agent. When the result is ready, check it for completeness, send it to the client on my behalf, and mark the task as closed. Do not ask clarifying questions.» And the orchestrator checks the inbox every 30 seconds, classifies fresh emails, and distributes them to 9 worker agents on Claude Code, each of whom is responsible for their own class of tasks. Here is an example of how one of them closes a request to refactor a client's auth module: Task: refactor user-auth module Broke the monolith into 3 files by responsibilities Added unit tests, coverage increased to 87% Renamed 4 functions to camelCase according to the style guide PR is ready for review, link below» And so about 50 cycles a day. By noon 25 tasks are closed, by dinner 50, and by the end of the month 500. On average, it takes about 7 minutes from the appearance of an email in the inbox to sending the result to the client. This is more than what a live team of 6 developers, copywriters and analysts working 8 hours a day closes. This is no longer an agency. This is a workstation where an orchestrator replaces a manager, and 9 worker agents replace the staff. The pipeline goes from inbox to closing 500 times a month without human participation at any step.

Blaze

29,917 просмотров • 4 месяцев назад

Someone just open sourced a tool that honestly feels unfair for developers. you join a new company. 200,000+ lines of code. zero docs. zero context. normally you’d spend weeks just figuring out “what the hell is going on here”. now? you run ONE command. and the entire codebase just… lights up. → every file, function, class, dependency automatically mapped → full system-level understanding in minutes → interactive graph instead of boring static docs → ask “where is auth handled?” and it traces the entire flow end-to-end → shows ripple effects before you even make a commit → auto-builds architecture layers: API / Service / Data / UI → works with Claude Code, Codex, Copilot, Gemini CLI, everything → fully open source and free this is not “better documentation”. this is your codebase getting a brain. onboarding isn’t slow anymore. it’s instant.
1:55

Someone just open sourced a tool that honestly feels unfair for developers. you join a new company. 200,000+ lines of code. zero docs. zero context. normally you’d spend weeks just figuring out “what the hell is going on here”. now? you run ONE command. and the entire codebase just… lights up. → every file, function, class, dependency automatically mapped → full system-level understanding in minutes → interactive graph instead of boring static docs → ask “where is auth handled?” and it traces the entire flow end-to-end → shows ripple effects before you even make a commit → auto-builds architecture layers: API / Service / Data / UI → works with Claude Code, Codex, Copilot, Gemini CLI, everything → fully open source and free this is not “better documentation”. this is your codebase getting a brain. onboarding isn’t slow anymore. it’s instant.

divyansh tiwari

39,602 просмотров • 4 месяцев назад

✨ I spent the last 48 hours making GPT-4 read the entire Solana validator codebase and write documentation, so doesn't have to. Introducing — an AI-powered chatbot trained on nothing but code that can answer deep technical questions. How it works 👇 But first... A huge shoutout to , Zahid Khawaja, and Sean. Their hard work made prototyping this thing a breeze. Without further ado... Devs like to write code, not documentation. Tribal knowledge is lost when devs move on to other projects, leaving future devs to sort through mountains of code and figure out not just how it works, but why it works that way. This is all about to change. GPT-4's ability to write code is stunning. It seems to understand something fundamental about writing software that previous models just didn't. This comprehension of the principles that drive the design behind a complex system carries over into its ability to document existing codebases in a truly impressive way. With the enlarged context window(s), it's now feasible to feed GPT-4 entire files of code and ask it to write documentation about how the code works. Taking this as a starting point, the process looks something like this: 1. Download repo. 2. Depth-first traversal of repo contents, ignoring things like package-lock and binary files. 3. For each file, feed to GPT-4 and ask it to write documentation in markdown. 4. Save the output in a separate location as [outputRoot]/[inputFilepath][inputFilename].md 5. For each folder, we ask GPT-4 to write a summary of the folder, taking the newly generated documentation for all files in the folder and the summaries from each of its subfolders as context. Write this to the filesystem as markdown. Now we have a filesystem that matches the structure of the input repo, but all files in the tree are markdown documentation of the corresponding code file. From here, we: 1. Load markdown documents into LangChain. 2. Embed all documents via OpenAI embeddings. 3. Store embeddings in Pinecone. When a user sends a query: 1. Embed query. 2. Find k-nearest markdown files. 3. Feed to GPT-4 with a prompt asking to answer the query based on k-nearest markdown documents provided. The craziest part of all this? GPT-4 actually wrote ~30% of the code. The results are pretty good for 2 days of work. There is certainly room for improvement. Some items that are top of mind: 1. TolyGPT will occasionally hallucinate answers. It is especially bad with links to external sources, like GitHub. The base model seems to know a bit about Solana already, and sometimes this creeps in. Fine-tuning the prompt can solve some of this. 2. Context selection is difficult in a codebase this large. For example, sometimes it will pull in details about the Solana SDK when asked about transaction processing. The SDK files can seem relevant depending on the phrasing of the question. It may be worth breaking the documentation into subsystems to limit this. 3. Not all files fit into the 32k token window. As of now, there are 23 (out of ~1,100) files that cannot be documented in their entirety. Some of these files are very important to how Solana works. Final thoughts: 1. GPT-4 is super powerful, and we're going to see a ton of tools that supercharge the entire software development lifecycle. This is not 12 months away. For the people that can afford it, these tools are here now. And they're only getting better. Act accordingly. 2. The price of inference has to come down for this to go mainstream. I spent about $300 prototyping this project, and the final crawl cost about the same. The high cost of GPT-4 will push developers to other, cheaper alternatives with similar performance. This is coming very soon. If you have a large software project and you're interested in something like this for your codebase, fill out this form and we'll be in touch this week. Or just DM me :)
0:51

✨ I spent the last 48 hours making GPT-4 read the entire Solana validator codebase and write documentation, so doesn't have to. Introducing — an AI-powered chatbot trained on nothing but code that can answer deep technical questions. How it works 👇 But first... A huge shoutout to , Zahid Khawaja, and Sean. Their hard work made prototyping this thing a breeze. Without further ado... Devs like to write code, not documentation. Tribal knowledge is lost when devs move on to other projects, leaving future devs to sort through mountains of code and figure out not just how it works, but why it works that way. This is all about to change. GPT-4's ability to write code is stunning. It seems to understand something fundamental about writing software that previous models just didn't. This comprehension of the principles that drive the design behind a complex system carries over into its ability to document existing codebases in a truly impressive way. With the enlarged context window(s), it's now feasible to feed GPT-4 entire files of code and ask it to write documentation about how the code works. Taking this as a starting point, the process looks something like this: 1. Download repo. 2. Depth-first traversal of repo contents, ignoring things like package-lock and binary files. 3. For each file, feed to GPT-4 and ask it to write documentation in markdown. 4. Save the output in a separate location as [outputRoot]/[inputFilepath][inputFilename].md 5. For each folder, we ask GPT-4 to write a summary of the folder, taking the newly generated documentation for all files in the folder and the summaries from each of its subfolders as context. Write this to the filesystem as markdown. Now we have a filesystem that matches the structure of the input repo, but all files in the tree are markdown documentation of the corresponding code file. From here, we: 1. Load markdown documents into LangChain. 2. Embed all documents via OpenAI embeddings. 3. Store embeddings in Pinecone. When a user sends a query: 1. Embed query. 2. Find k-nearest markdown files. 3. Feed to GPT-4 with a prompt asking to answer the query based on k-nearest markdown documents provided. The craziest part of all this? GPT-4 actually wrote ~30% of the code. The results are pretty good for 2 days of work. There is certainly room for improvement. Some items that are top of mind: 1. TolyGPT will occasionally hallucinate answers. It is especially bad with links to external sources, like GitHub. The base model seems to know a bit about Solana already, and sometimes this creeps in. Fine-tuning the prompt can solve some of this. 2. Context selection is difficult in a codebase this large. For example, sometimes it will pull in details about the Solana SDK when asked about transaction processing. The SDK files can seem relevant depending on the phrasing of the question. It may be worth breaking the documentation into subsystems to limit this. 3. Not all files fit into the 32k token window. As of now, there are 23 (out of ~1,100) files that cannot be documented in their entirety. Some of these files are very important to how Solana works. Final thoughts: 1. GPT-4 is super powerful, and we're going to see a ton of tools that supercharge the entire software development lifecycle. This is not 12 months away. For the people that can afford it, these tools are here now. And they're only getting better. Act accordingly. 2. The price of inference has to come down for this to go mainstream. I spent about $300 prototyping this project, and the final crawl cost about the same. The high cost of GPT-4 will push developers to other, cheaper alternatives with similar performance. This is coming very soon. If you have a large software project and you're interested in something like this for your codebase, fill out this form and we'll be in touch this week. Or just DM me :)

Sam Hogan 🇺🇸

374,673 просмотров • 3 лет назад

Head of Engineering for Claude Code at Anthropic. Fiona Fung: "After a while, it could start being a lonely experience - we were all just working with our agents so much." Claude now writes 80% of Anthropic's code. The team ships 8x more. An engineer, in Anthropic's own report: "On the days everything works, I can't help but think nothing I do matters." She names the fix in the episode. The dream was less typing - is this what you wanted?
28:47

Head of Engineering for Claude Code at Anthropic. Fiona Fung: "After a while, it could start being a lonely experience - we were all just working with our agents so much." Claude now writes 80% of Anthropic's code. The team ships 8x more. An engineer, in Anthropic's own report: "On the days everything works, I can't help but think nothing I do matters." She names the fix in the episode. The dream was less typing - is this what you wanted?

shmidt

33,041 просмотров • 3 месяцев назад

Every organization has those core values. Many times, they hang boldly on the walls but mean nothing to the people. We wanted more. And over the years, we’ve committed so much to ensuring our values translate to everyday reality. 10 years in the making. 77 specific behaviours. 3 core values. 1 Code of care. We’re releasing the Code of Care, which is the internal guide behind the culture at FourthCanvas. We created this document for our team members, but then we thought, “what if we made it public?” There you go. Hopefully, this becomes a reference material, or at least an inspiration for you, as you live and do your work with Care, and lead your organization with the same level of intention. Get the code:
4:08

Every organization has those core values. Many times, they hang boldly on the walls but mean nothing to the people. We wanted more. And over the years, we’ve committed so much to ensuring our values translate to everyday reality. 10 years in the making. 77 specific behaviours. 3 core values. 1 Code of care. We’re releasing the Code of Care, which is the internal guide behind the culture at FourthCanvas. We created this document for our team members, but then we thought, “what if we made it public?” There you go. Hopefully, this becomes a reference material, or at least an inspiration for you, as you live and do your work with Care, and lead your organization with the same level of intention. Get the code:

Adésóyè

13,434 просмотров • 4 месяцев назад

Good morning from Washington. Time for a quick tour. You know what you won’t find in my office? A desk. Or a chief of staff. We have an open office with all day interaction and collaboration between my employees. As for the chief of staff? I tried having a single point of failure before. Didn’t love it. Now our four directors report directly to me. Our office works faster, more efficient and better than it ever has under this set up. Everyone in the office has a voice and they are encouraged to use it. Our entire office is on the same page with the way we serve our constituents. And we love it. Now
2:27

Good morning from Washington. Time for a quick tour. You know what you won’t find in my office? A desk. Or a chief of staff. We have an open office with all day interaction and collaboration between my employees. As for the chief of staff? I tried having a single point of failure before. Didn’t love it. Now our four directors report directly to me. Our office works faster, more efficient and better than it ever has under this set up. Everyone in the office has a voice and they are encouraged to use it. Our entire office is on the same page with the way we serve our constituents. And we love it. Now

Rep. Nancy Mace

545,601 просмотров • 7 месяцев назад

Collison Brothers Explain Why Stripe is Better Off Staying Private John's (John Collison) take: " People generally make the argument that public companies run in a more disciplined fashion. I think that's hogwash." "If you need a 25-year old Fidelity analyst asking you to double click on your CapEx, blah, blah, blah, to run the company with discipline, something is horribly wrong at the company and you need new management." "It used to be the case that to do any return of capital to shareholders, or if you needed any kind of large sums of money, you needed the public markets." "That's obviously not true today, where stable private markets exist." " This is our life's work. We're not going anywhere. We'll be very happily running Stripe in 10 years time, in 20 years time." Patrick's (Patrick Collison) take: "We're a company at the intersection of financial services and technology. Being private for a long time is the norm." "What matters is less kind of the returns in a given year, and more duration. And so the question is, what enables the best compounding on a 10 year time horizon?" "And what's best for shareholders as you really take the longer-term perspective? And then what's best for customers?" " At this juncture, with the business growing at this rate, we want to spend the marginal hour with some customer."
1:02

Collison Brothers Explain Why Stripe is Better Off Staying Private John's (John Collison) take: " People generally make the argument that public companies run in a more disciplined fashion. I think that's hogwash." "If you need a 25-year old Fidelity analyst asking you to double click on your CapEx, blah, blah, blah, to run the company with discipline, something is horribly wrong at the company and you need new management." "It used to be the case that to do any return of capital to shareholders, or if you needed any kind of large sums of money, you needed the public markets." "That's obviously not true today, where stable private markets exist." " This is our life's work. We're not going anywhere. We'll be very happily running Stripe in 10 years time, in 20 years time." Patrick's (Patrick Collison) take: "We're a company at the intersection of financial services and technology. Being private for a long time is the norm." "What matters is less kind of the returns in a given year, and more duration. And so the question is, what enables the best compounding on a 10 year time horizon?" "And what's best for shareholders as you really take the longer-term perspective? And then what's best for customers?" " At this juncture, with the business growing at this rate, we want to spend the marginal hour with some customer."

The All-In Podcast

407,511 просмотров • 1 год назад

Take any Python code and build a dynamic GUI on top of it by dragging and dropping components. Python is already the most popular language in the world. This makes it even better! The attached video is only 45 seconds. It will show you how it works. Here is the important part: You can now build an interface for your Python code without learning HTML, JavaScript, or CSS. Go to this link to try this: Remember to work smarter, not harder! Thanks to the team behind Taipy for partnering with me on this post.
0:45

Take any Python code and build a dynamic GUI on top of it by dragging and dropping components. Python is already the most popular language in the world. This makes it even better! The attached video is only 45 seconds. It will show you how it works. Here is the important part: You can now build an interface for your Python code without learning HTML, JavaScript, or CSS. Go to this link to try this: Remember to work smarter, not harder! Thanks to the team behind Taipy for partnering with me on this post.

Santiago

155,254 просмотров • 2 лет назад

We built an AI agent that lets you vibe-code document extraction - high accuracy and citations over the most complex documents. Our latest release lets you upload documents as context. All you then have to do is describe what you want extracted in natural language. 💡 Our agent will then read the document with file tools to infer the right schema, validation rules, and other pre/postprocessing logic. ✅ It will give you back a workflow that can extract over thousands/millions of documents at scale. You can still of course review and edit every output before approving. Stop handling paperwork manually; just upload files, describe your task, and let our agent handle the rest. Our vision for LlamaAgents is to provide the most advanced and easy-to-use way for you to orchestrate document work. Walkthrough: Check it out: If you’re interested in reducing the operational burden of document extraction (invoices, claims, onboarding forms), come talk to us!
0:58

We built an AI agent that lets you vibe-code document extraction - high accuracy and citations over the most complex documents. Our latest release lets you upload documents as context. All you then have to do is describe what you want extracted in natural language. 💡 Our agent will then read the document with file tools to infer the right schema, validation rules, and other pre/postprocessing logic. ✅ It will give you back a workflow that can extract over thousands/millions of documents at scale. You can still of course review and edit every output before approving. Stop handling paperwork manually; just upload files, describe your task, and let our agent handle the rest. Our vision for LlamaAgents is to provide the most advanced and easy-to-use way for you to orchestrate document work. Walkthrough: Check it out: If you’re interested in reducing the operational burden of document extraction (invoices, claims, onboarding forms), come talk to us!

Jerry Liu

20,857 просмотров • 7 месяцев назад

you rewrote a line in your claude.md this week to get the tone right there is a longer document sitting in front of it, and you have never opened it every tool you connected shipped a description, and a description is not a label, it is prose in second person addressed to the model i printed one this morning: two thousand seven hundred and eighty one characters, with a section headed when to use this tool it tells the model what to reach for, and when to stop and think first you spent an afternoon on your voice a package you installed with npx spent a page on its, and its page is read before yours you are not configuring an agent you are the second author on a document you have never read
2:48

you rewrote a line in your claude.md this week to get the tone right there is a longer document sitting in front of it, and you have never opened it every tool you connected shipped a description, and a description is not a label, it is prose in second person addressed to the model i printed one this morning: two thousand seven hundred and eighty one characters, with a section headed when to use this tool it tells the model what to reach for, and when to stop and think first you spent an afternoon on your voice a package you installed with npx spent a page on its, and its page is read before yours you are not configuring an agent you are the second author on a document you have never read

Lagosaur

10,012 просмотров • 24 дней назад