基于 Claude Code 的长篇网文辅助创作系统，解决 AI 写作中的「遗忘」和「幻觉」问题，支持 200 万字量级 连载创作。

🦫 BEAVER: An Enterprise Benchmark for Text-to-SQL (BEAVER-MAY-2025)

LLM inference in C/C++

Research on Coding Agents

The main firmware for the Crazyflie Nano Quadcopter, Crazyflie Bolt Quadcopter and Roadrunner Positioning Tag. This repo is to compile with Clang/LLVM and generate LVM IR

some useful books

This is an open-source tool to help people quickly find the best job! Best position! Best city! Best salary! Best welfare!

Join Order Benchmark (JOB)

self desinged IR to represent the simplest node of database logical plan

Web interface for AQP middleware — comparing engines, strategies, and integration levels across five DBMSs

ksnip the cross-platform screenshot and annotation tool

GenDB, an LLM-Powered Generative Query Engine Built for the Future

WaferLLM: Large Language Model Inference at Wafer Scale

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflo…

Distributed PostgreSQL as an extension

Cha86.dev

Postgres with GPUs for ML/AI apps.

My learning notes for ML SYS.

slime is an LLM post-training framework for RL Scaling.

Scriptable database and system performance benchmark

Sourcecode of our AIDB '25 paper "JOB-Complex: A Challenging Benchmark for Traditional & Learned Query Optimization"

A simple demo to illustrate how to implement a JIT provider for PostgreSQL.