SilenNaihin/screenpipe
YC (S26) | Open Computer History | Record your screen continuously locally and provide context to your agents (Claude, Codex, Openclaw, Hermes, Runner...)
AI for new knowledge.
YC (S26) | Open Computer History | Record your screen continuously locally and provide context to your agents (Claude, Codex, Openclaw, Hermes, Runner...)
๐ฅ Enterprise SaaS Starter Kit - Kickstart your enterprise app development with the Next.js SaaS boilerplate ๐
Interactive ML articles with scroll-synced animations and Jupyter notebooks.
SAEs waste 74% of their features, cosine scoring finds the missing ones
GPU fleet monitoring dashboard
Continual Learning as a Service
A high-throughput and memory-efficient inference and serving engine for LLMs
ETL for Berkeley alab experimental data to MPContribs. Goal is to accelerate experimentation and replicability for fully autonomous material lab.
AI agents running research on single-GPU nanochat training automatically
The best ChatGPT that $100 can buy.
NanoGPT (124M) in 2 minutes
The simplest, fastest repository for training/finetuning medium-sized GPTs.
Your own personal AI assistant. Any OS. Any Platform. The lobster way. ๐ฆ
https://blog.silennai.com/genetic-algorithm
Open-source agentic job processing pipeline - scraping, enrichment, and workflow automation
Genome modeling and design across all domains of life
Anthropic's original performance take-home, now open for you to try!
Article: https://medium.com/@silen.naihin/tic-tac-toe-but-more-confusing-and-way-cooler-1a3db38bd107, Video: https://www.youtube.com/watch?v=VzxxwZ8d7Q4&t=4s
Platform for materials scientists to contribute and disseminate their materials data through Materials Project
Specify what you want it to build, the AI asks for clarification, and then builds it.
Repo of books I read or did my best to read while learning quantum computing.
Designed to streamline and simplify your AI workflows, it offers Python typing support, error checking, and a smart usage meter to manage API costs efficiently. Plus, it features built-in examples with ChromaDB and efficient tools for prompt engineering.