🚀 Introducing Structured RAG (S-RAG) S-RAG transforms unstructured data into a structured, query-aware representation. It then uses formal queries over structured data at runtime, so AI21 Maestro can retrieve accurate values, ensure completeness, handle inconsistencies, and

This is really fun - here are my notes on how to try it out using Docker for Mac simonwillison.net/2025/Nov/12/h4…

terminal browser to trowse without distractions lightweight hyper client & crossterm (rust) already had majority of the libraries ready from the search engine links are categorized (intern/extern/subdomain) and files (pdf, image, etc) berkeley font, ghosty term

12 papers already published in good psychological journals were resubmitted with fake author names and non-prestigious institutional affiliations. 9 made it through the submission process undetected. 8 were rejected.

I've been learning about agent memory for the past few weeks. This new blog summarizes everything I've learned so far: • What is agent memory, and why do you need it • What are the types of memory (and what categorization approaches are there?) • How do you manage memory in

Unsloth has a great guide on fine tuning LLMs. > selecting the right model > preparing your own data > understanding quantisation, lora > training params & hyperparams > evaluation An SLM fine tuned for a specific task often shows better results than using a generic LLM. Fine

Code: github.com/virattt/dexter

Grab the step by step Python guide here: pyquantnews.com/the-pyquant-ne…

Secret of how hedge funds reduce their risk of loss: Kelly Criterion Here's how:

Wow, such a great open-source drop! Put Claude on steroids with 120+ scientific skills spanning maths, biology, chemistry, medicine, engineering, & more 🤯 If you’ve ever wanted Claude to act like a research assistant, this gets very close. Free and open source. Repo in 🧵↓

🎉 DeepCode hits 10.9k stars on GitHub! ⭐ (👉 github.com/HKUDS/DeepCode) Huge thanks to our amazing open source community! 🙏 New Release Features: - 🏆 First Vibe Coding to Beat Human Experts - Outperforms ML PhDs from top universities (75.9% vs 72.4%) and Cursor & Claude Code

Banger paper for agent builders. Multi-agent systems often underdeliver. The problem isn't how the agents themselves are built. It's how they're organized. They are mostly built with fixed chains, trees, and graphs that can't adapt as tasks evolve. But what if the system could

AI agents can't learn from experience. Until now. Earlier this year, one of our customers asked us: "If your web agent uses my website once, will it be easier the next time it visits?" The question highlighted a fundamental gap between human intelligence and AI agents: Humans

Post and code: "Separating Signal from Noise: A Practical Guide to Evaluating Alpha Factors with Alphalens [Python Package]" medium.com/@er.mananjain2… & github.com/quantopian/alp…

this russian guy found a way to learn anything 10x faster🚨 saved 1,460h NotebookLM + Gemini + Obsidian dumps any source > AI strips duplicates > keeps only what u don't know yet 20 YouTube videos on the same topic each one repeats the same 20% of information this bundle

Cool

this Obsidian + AI is the new hot combo. few people know that the CEO of Obsidian kepano has made multiples skills for Claude Code and Codex that you can use right now both for your codebase and your personal vault.

Converts PDFs to Markdown at 100 pages per second on CPUs github.com/opendataloader…

We open sourced an operating system for ai agents 137k lines of rust, MIT licensed we love OpenClaw🦞 and it inspired a lot of what we built. but we wanted something that works at the kernel level so we built OpenFang agents run inside WASM sandboxes the same way processes

Google DeepMind just used AlphaEvolve to breed entirely new game-theory algorithms that outperform ones humans spent years designing the discovered algorithms use mechanisms so non-intuitive that no human researcher would have tried them. here's what actually happened and why