GitSpot @gitspot

@deedydas 🚨 Google DeepMind new paper: Video models are zero-shot learners and reasoners

keyboard_arrow_left Previous keyboard_arrow_right Next

AI Agents Stack amzn.to/4mlbsDA

>be India >spend just ₹615 crore on the Chandrayaan-3 mission >spend ₹807 cr on a 3.75 km stretch of an ugly ropeway

How to build an AI agent

holy crap this paper might actually fix vibe coding keeping a repo’s structure in context is a real struggle, and they’ve cooked up a graph-guided code generation framework that could solve that for good according to the authors, the framework is coming out soon

I think about this at least 1x a week.

a senior engineer at google just dropped a 400-page free book on docs for review: agentic design patterns. the table of contents looks like everything you need to know about agents + code: > advanced prompt techniques > multi-agent patterns > tool use and MCP > you name it

Our paper on autonomous scientific research is accepted to Findings of #EMNLP2025! 🎉 We introduce Agent Laboratory, a framework that accelerates scientific discovery by teaming human researchers with LLM agents.

a new Agentic model that can run on a single consumer GPU at home: ByteDance Seed OSS 36B > very strong at coding > excellent at multi-turn tool calling & Agentic tasks > 500k context window as i have been saying, Bytedance is Tier S

At MIT, I learned about RNNs in my NLP class with Prof. Michael Collins. He built a model from my keystrokes to predict who I was. To me, it felt like a magic box. Years later, when I had to teach RNNs, I forced myself to go inside the box. ⬇️ Download: byhand.ai/rnn…

A banger on how to publish the ultimate journal approved research paper in 2025:

okay interesting discovery: if 17 creators post identical content within 3 minutes the algorithm thinks it's "breaking news" pushes everything to trending tested this 32 times with a few in-house ghost creators ALWAYS works manufacturing news > reporting news

Chain-of-Agents Interesting idea to train a single model with the capabilities of a multi-agent system. 84.6% reduction in inference cost! Distillation and Agentic RL are no joke! Here are my notes:

So Alibaba Qwen has released the best image editing model... 100% open source! You can edit any photo using natural language. It can be used both locally and online. More below

My favorite deep learning intuition is that neural net layers are a series of geometric transforms

I spent my summer building TinyTPU : An open source ML inference and training chip. it can do end to end inference + training ENTIRELY on chip. here's how I did it👇:

this is one of the most remarkable technical blog posts I’ve ever read

Traditional RAG vs. Agentic RAG, clearly explained (with visuals):

Starting today, GPT-5 mini is available in @code! With a 0x multiplier, GPT-5 mini is a lightweight version of OpenAI's most advanced model. Try it out and let us know what you think!

I think about this post once a month