On the Theoretical Limitations of Embedding-Based Retrieval

LLM KV cache compression made easy

Frontier: A Discrete-Event Simulator for Modern LLM Serving

A novel embedding training algorithm leveraging ANN search and achieved SOTA retrieval on Trec DL 2019 and OpenQA benchmarks

A tutorial on modern GPU programming for machine learning systems

FoundationDB - the open source, distributed, transactional key-value store

Efficient Long-Context LLM Serving with Head-Aware KV Reuse and SegPagedAttention

DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms

AI 时代的伯克希尔：基于 Claude Code / Codex 的价值投资研究框架。巴菲特·芒格·段永平·李录四大师方法论 + 多Agent并行研究。| AI-era Berkshire: a value investing research framework built for Claude Code / Codex. 4 masters' methodologies + multi…

Unlimited OCR Works: Welcome the Era of One-shot Long-horizon Parsing.

System for AI Education Resource.

🎙️ A 0.1B Omni model trained from scratch, capable of listening, speaking, and seeing!

high-performance inference and serving library for interactive autoregressive video and world models

Converts text to speech in realtime

C++-based high-performance parallel environment execution engine (vectorized env) for general RL environments.

Graphs that teach > graphs that impress. Turn any code into an interactive knowledge graph you can explore, search, and ask questions about. Works with Claude Code, Codex, Cursor, Copilot, Gemini C…

Techniques and numbers for estimating system's performance from first-principles

Kimi Code CLI — The Starting Point for Next-Gen Agents

Moshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec.

An end-to-end open ecosystem for robot learning

🚀 Awesome System for Machine Learning ⚡️ AI System Papers and Industry Practice. ⚡️ System for Machine Learning, LLM (Large Language Model), GenAI (Generative AI). 🍻 OSDI, NSDI, SIGCOMM, SoCC, MLSy…

Research on Coding Agents

Claude Code skill: Generate file-by-file code tutorial websites for any repository with parallel agent teams

A minimal, end-to-end event-driven trading system written in Rust, designed to model market data ingestion, pluggable strategies, and deterministic execution.

The open-source app everyone uses to manage agents at work

from vibe coding to agentic engineering - practice makes claude perfect

PM Skills Marketplace: 100+ agentic skills, commands, and plugins — from discovery to strategy, execution, launch, and growth.

🧠 Make your agents learn from experience. Now available as a hosted solution at kayba.ai