yi 🛡️ (@drtwo101) 's Twitter Profile
yi 🛡️

@drtwo101

Building @codatta_io which transforms foundational models into vertical AI solutions. prev: ex-(Pinterest, Alipay), PhD in ML.

ID: 41297329

calendar_today20-05-2009 04:37:59

1,1K Tweet

7,7K Followers

1,1K Following

Santiago (@svpino) 's Twitter Profile Photo

Really good book for learning how to build agentic systems! If you want to move beyond simple AI pipelines and get into agents, this will be a great starting point. The book is highly practical and will show you how to build autonomous agents that can plan, reason, use tools,

Really good book for learning how to build agentic systems!

If you want to move beyond simple AI pipelines and get into agents, this will be a great starting point.

The book is highly practical and will show you how to build autonomous agents that can plan, reason, use tools,
Andrej Karpathy (@karpathy) 's Twitter Profile Photo

Love this project: nanoGPT -> recursive self-improvement benchmark. Good old nanoGPT keeps on giving and surprising :) - First I wrote it as a small little repo to teach people the basics of training GPTs. - Then it became a target and baseline for my port to direct C/CUDA

Google AI Developers (@googleaidevs) 's Twitter Profile Photo

Our latest Gemma 3n model pushes the boundaries of multimodal AI, natively understanding and processing information from images, audio, and more. See Gemma 3n in action 🧵↓

Min Choi (@minchoi) 's Twitter Profile Photo

Claude Code has really changed the vibe coding game. People are coming up with wild use cases. There's a shift. 10 examples: 1. Prompt: Create an amazing animation multicolor and interactive using p5js

Ivan Fioravanti ᯅ (@ivanfioravanti) 's Twitter Profile Photo

DeepSeek-R1-0528-5bit on MLX pushing M3 Ultra 512GB to its limits! 501GB used mem visibile on mactop in the video! Context: 4K tokens Prompt: 190.29 t/s Gen: 11.37 t/s Peak Mem: 487.48 GB! THIS IS APPLE MLX!

Luis (@lusxvr) 's Twitter Profile Photo

Today, we are open-sourcing our pipeline to deduplicate large-scale image datasets. On one GPU, we can deduplicate 10k images against 1M indexed test images in ~60 seconds. But how?

Today, we are open-sourcing our pipeline to deduplicate large-scale image datasets. 

On one GPU, we can deduplicate 10k images against 1M indexed test images in ~60 seconds. But how?
dax (@thdxr) 's Twitter Profile Photo

it's funny how many people wrote up huge predictions for MCP without even looking into how LLM performance degrades when you add even 10 tools to them

AI Will (@financeyf5) 's Twitter Profile Photo

这是一本精彩、视觉震撼的深度学习免费入门书籍。 强烈推荐给那些想了解基本概况的好奇人群。

这是一本精彩、视觉震撼的深度学习免费入门书籍。

强烈推荐给那些想了解基本概况的好奇人群。