Loading video...

Video Failed to Load

Go Home

Finally! A RAG over code solution that actually works (open-source). Naive chunking used in RAG isn't suited for code. This is because codebases have long-range dependencies, cross-file references, etc., that independent text chunks just can't capture. Graph-Code is a graph-driven RAG system that solves this. It analyzes the Python...

_avichawla's profile picture

Avi Chawla

69,109 subscribers

121,819 views • 11 months ago •via X (Twitter)

Health & WellnessScience & TechnologyEducation
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Code reviews Codex goes beyond static analysis—it checks a PR against its intent, reasons across the codebase and dependencies, and can run code to validate the behavior of changes. Set it up to auto-review new PRs in your repo, or tag @ codex.
0:13

Code reviews Codex goes beyond static analysis—it checks a PR against its intent, reasons across the codebase and dependencies, and can run code to validate the behavior of changes. Set it up to auto-review new PRs in your repo, or tag @ codex.

OpenAI Developers

59,172 views • 9 months ago

A few of you asked for the code for these SwiftUI prototypes, I just added them to my open source repo, link in the replies
0:10

A few of you asked for the code for these SwiftUI prototypes, I just added them to my open source repo, link in the replies

Mike Lee

70,047 views • 1 year ago

Day 6 of Stealing Designs and converting them into Code I have made the code for the Dashboard open source Check this out in replies
0:14

Day 6 of Stealing Designs and converting them into Code I have made the code for the Dashboard open source Check this out in replies

Karan Kendre

21,958 views • 5 months ago

My project has 39,205 lines of code, and Cursor can't answer questions about it. Cursor's context seems to be capped at around 10,000 tokens. Unfortunately, this is not enough for any decent-sized project. If you have a large codebase, check out Augment Code. This thing is faaaast! I'm currently using their Visual Studio Code plugin, but you can also use them on JetBrains, Neovim, and even Vim. (I'm a Neovim fan, but Copilot's implementation for Neovim is nowhere as good as Augment Code.) Augment Code was gracious enough to sponsor this post. After you install their extension and run it for the first time, it will index your entire codebase. This is why it can answer questions as fast as it does, regardless of the size of your codebase. Augment Code supports chat and completions like every other AI coding assistant, but its killer feature is "Next Edit." When you make a change, two things happen: 1. The model analyzes the change to determine the ripple effects across your *entire* codebase. 2. The model suggests everything you need to update to ensure everything works correctly. This is pretty wild!
0:19

My project has 39,205 lines of code, and Cursor can't answer questions about it. Cursor's context seems to be capped at around 10,000 tokens. Unfortunately, this is not enough for any decent-sized project. If you have a large codebase, check out Augment Code. This thing is faaaast! I'm currently using their Visual Studio Code plugin, but you can also use them on JetBrains, Neovim, and even Vim. (I'm a Neovim fan, but Copilot's implementation for Neovim is nowhere as good as Augment Code.) Augment Code was gracious enough to sponsor this post. After you install their extension and run it for the first time, it will index your entire codebase. This is why it can answer questions as fast as it does, regardless of the size of your codebase. Augment Code supports chat and completions like every other AI coding assistant, but its killer feature is "Next Edit." When you make a change, two things happen: 1. The model analyzes the change to determine the ripple effects across your *entire* codebase. 2. The model suggests everything you need to update to ensure everything works correctly. This is pretty wild!

Santiago

247,775 views • 1 year ago

No more excuses to not help fund the open source software that powers your code Just type /tribute in Claude Code
0:19

No more excuses to not help fund the open source software that powers your code Just type /tribute in Claude Code

Josh Cohenzadeh

62,299 views • 4 months ago

Don't trust, Verify ! You can now dig into the core building blocks of Keet to verify the code and cryptography that is being used ! Visit & to view the first wave of our open-source code ! More updates and code to follow soon!🍐
0:15

Don't trust, Verify ! You can now dig into the core building blocks of Keet to verify the code and cryptography that is being used ! Visit & to view the first wave of our open-source code ! More updates and code to follow soon!🍐

Holepunch🕳🥊

140,249 views • 3 years ago

Seeing this post from Zach inspired me to explore some eye tracking experiments, shared these before but consolidated them into a post, shared the code in my open source repo for the eye drawing experiment, let me know if anyone wants the code to the other experiments, link in replies
0:14

Seeing this post from Zach inspired me to explore some eye tracking experiments, shared these before but consolidated them into a post, shared the code in my open source repo for the eye drawing experiment, let me know if anyone wants the code to the other experiments, link in replies

Mike Lee

27,314 views • 6 months ago

Building RAG is easy. Parsing real, unstructured data is the hard part. Most tools fail when documents get complicated. RAGFlow by InfiniFlow makes the entire process visual and flawless 🔥 It is an (open-source!) engine built specifically to find the exact needle in a data haystack, even across literally unlimited tokens. The platform comes packed with: → "Quality in, quality out" parsing for highly complex formats → Multiple recall paired with fused re-ranking → A built-in Python and JavaScript code executor for agents → An orchestrable ingestion pipeline Here's why it stands out: 1️⃣ Structural Understanding Instead of just scraping text, it handles tables across pages, scanned copies, slides, and Excel sheets natively using deep document understanding. 2️⃣ Grounded Citations Every answer is verifiable. The UI highlights the exact chunks used, allowing you to trace any response directly back to the source material. 3️⃣ Enterprise Synchronization Keep your context constantly updated with native data sync from Google Drive, Notion, Discord, and Confluence. Stop letting bad document parsing ruin your RAG systems. Best part? It's 100% Free and open-source. Link to the repo in 🧵↓
0:18

Building RAG is easy. Parsing real, unstructured data is the hard part. Most tools fail when documents get complicated. RAGFlow by InfiniFlow makes the entire process visual and flawless 🔥 It is an (open-source!) engine built specifically to find the exact needle in a data haystack, even across literally unlimited tokens. The platform comes packed with: → "Quality in, quality out" parsing for highly complex formats → Multiple recall paired with fused re-ranking → A built-in Python and JavaScript code executor for agents → An orchestrable ingestion pipeline Here's why it stands out: 1️⃣ Structural Understanding Instead of just scraping text, it handles tables across pages, scanned copies, slides, and Excel sheets natively using deep document understanding. 2️⃣ Grounded Citations Every answer is verifiable. The UI highlights the exact chunks used, allowing you to trace any response directly back to the source material. 3️⃣ Enterprise Synchronization Keep your context constantly updated with native data sync from Google Drive, Notion, Discord, and Confluence. Stop letting bad document parsing ruin your RAG systems. Best part? It's 100% Free and open-source. Link to the repo in 🧵↓

Charly Wargnier

19,131 views • 2 months ago

dog follows pointer + avoids collisions with no dependencies - builds the page as a graph, more subdivisions towards the text for perf - dijkstra shortest path
0:10

dog follows pointer + avoids collisions with no dependencies - builds the page as a graph, more subdivisions towards the text for perf - dijkstra shortest path

JohnPhamous

48,995 views • 3 months ago

GDScript Formatter can reorder your GDScript code for you! Signals and variables first, then built-in Godot functions, then public and private custom functions, etc. It follows the official style guide. It's open source, written in Rust, and runs fast. Give it a try:
0:18

GDScript Formatter can reorder your GDScript code for you! Signals and variables first, then built-in Godot functions, then public and private custom functions, etc. It follows the official style guide. It's open source, written in Rust, and runs fast. Give it a try:

GDQuest

10,445 views • 8 months ago

The fact this was actually found in the Mario 64 Source code Is fucking unsettling
0:14

The fact this was actually found in the Mario 64 Source code Is fucking unsettling

Skoopy

3,373,602 views • 2 years ago

6. Explain code Give a link to Bard and ask it a question about a file. It'll be able to explain what it's for and how it works. Prompt → Can you explain what is the file agent py in this repo? (your link)
0:18

6. Explain code Give a link to Bard and ask it a question about a file. It'll be able to explain what it's for and how it works. Prompt → Can you explain what is the file agent py in this repo? (your link)

Paul Couvert

749,048 views • 3 years ago

ELON WANTS TO OPEN-SOURCE 𝕏 – ALL OF IT No leaks, no guesses – just raw code, wide open. Code Drop: * Full 𝕏 codebase could go public – algorithms, features, all of it * Devs could build custom versions, fix bugs, or fork their own 𝕏 * It’s radical transparency in an industry that hides everything * Risks? Sure. But the upside rewrites how platforms are built If this happens, it’s not just open-source – it’s open power. Source: Elon Musk
0:16

ELON WANTS TO OPEN-SOURCE 𝕏 – ALL OF IT No leaks, no guesses – just raw code, wide open. Code Drop: * Full 𝕏 codebase could go public – algorithms, features, all of it * Devs could build custom versions, fix bugs, or fork their own 𝕏 * It’s radical transparency in an industry that hides everything * Risks? Sure. But the upside rewrites how platforms are built If this happens, it’s not just open-source – it’s open power. Source: Elon Musk

Mario Nawfal

128,607 views • 5 months ago

Meta just announced that Code Llama was now free for both research and commercial. This might the strongest competitor to ChatGPT: ▸ Can generate, explain, and debug your code ▸ Handles input 100,000 tokens ▸ Free for research + commercial use ▸ Outperforms most open models ▸ Comes in 7B, 13B, and 34B ▸ Supports Python, C++, Java, PHP, Typescript (Javascript), C#, and Bash Available in: ▸ Foundation base models (Code Llama) ▸ Python specializations (Code Llama - Python), ▸ Instruction-following models (Code Llama - Instruct)
0:11

Meta just announced that Code Llama was now free for both research and commercial. This might the strongest competitor to ChatGPT: ▸ Can generate, explain, and debug your code ▸ Handles input 100,000 tokens ▸ Free for research + commercial use ▸ Outperforms most open models ▸ Comes in 7B, 13B, and 34B ▸ Supports Python, C++, Java, PHP, Typescript (Javascript), C#, and Bash Available in: ▸ Foundation base models (Code Llama) ▸ Python specializations (Code Llama - Python), ▸ Instruction-following models (Code Llama - Instruct)

Lior Alexander

196,380 views • 2 years ago

Wow. You can literally chat with any GitHub repo now! Just pop "talkto" in front of "github" in the URL, and BOOM, you're chatting with your codebase! 🤯 It launches a full chat UI where you can ask the code anything. 100% open-source. Repo link below ↓
0:11

Wow. You can literally chat with any GitHub repo now! Just pop "talkto" in front of "github" in the URL, and BOOM, you're chatting with your codebase! 🤯 It launches a full chat UI where you can ask the code anything. 100% open-source. Repo link below ↓

Charly Wargnier

15,055 views • 1 year ago

AI-generated QR code meets Augmented Reality 👀 This is a WebAR experience accessible to users by scanning an AI-generated QR code, which also functions as an image target for AR tracking. Try it out for yourself in thread:
0:17

AI-generated QR code meets Augmented Reality 👀 This is a WebAR experience accessible to users by scanning an AI-generated QR code, which also functions as an image target for AR tracking. Try it out for yourself in thread:

Ian Curtis

100,910 views • 3 years ago

It seems this isn't clicking to you yet... You can "vibe code" a side project. You can vibe code a quick landing page. You cannot vibe code enterprise infrastructure. Amazon just found out the hard way that letting an LLM guess your backend logic is a massive liability. The pendulum is about to swing aggressively back toward human oversight.
0:12

It seems this isn't clicking to you yet... You can "vibe code" a side project. You can vibe code a quick landing page. You cannot vibe code enterprise infrastructure. Amazon just found out the hard way that letting an LLM guess your backend logic is a massive liability. The pendulum is about to swing aggressively back toward human oversight.

Miye

66,996 views • 2 months ago

Next.js 15.1: Easier error debugging ✓ Hide¹ stack frames for dependency code ✓ Surface source of the error in your code Video: Terminal output² of a Zod parsing error
0:10

Next.js 15.1: Easier error debugging ✓ Hide¹ stack frames for dependency code ✓ Surface source of the error in your code Video: Terminal output² of a Zod parsing error

Delba

49,593 views • 1 year ago

The Terragon team has been cooking and has shipped some 🔥 ways to super charge Claude Code. First up, Automations. Schedule Claude Code to run at certain times of the day! I use it to get a summary of all of the changes that happened across the codebase every morning.
0:12

The Terragon team has been cooking and has shipped some 🔥 ways to super charge Claude Code. First up, Automations. Schedule Claude Code to run at certain times of the day! I use it to get a summary of all of the changes that happened across the codebase every morning.

Sawyer Hood

15,741 views • 10 months ago