Stars
Tracking operations using source locations in MLIR
Iterative agent harness improvement: run a coding agent on a hard task, generate the reusable tooling it was missing, qualify it, and replay fresh sessions with it activated. Works with Codex and C…
A curriculum for learning about gpu performance engineering, from scratch to what the frontier AI labs do
Free Weather Forecast API for non-commercial use
A open source reimplementation of Google's Tensor Processing Unit (TPU).
System Level Intelligent Router for Mixture-of-Models at Cloud, Data Center and Edge
A high-throughput and memory-efficient inference and serving engine for LLMs
APXM core runtime: AIS dialect, dekk CLI, vLLM-fork glue, lifecycle skills
Linux Memory Forensics Framework That Transforms Memory Dumps Into a Navigable Filesystem
Random program generation based on semantic reification (PLDI'26)
A high-performance, universal serving framework for any-to-any models.
a projected filesystem that turns APIs, services, and data sources into local files and directories
htop-like TUI for real-time RDMA network monitoring.
Omnigent is an open-source AI agent framework and meta-harness: orchestrate Claude Code, Codex, Cursor, Pi, and custom agents — swap harnesses without rewriting, enforce policies and sandboxing, an…
x86 JIT playground for documenting and exploring microarchitectural implementation details
Security scanner for AI agent skills. Detect vulnerabilities, malicious patterns, and security risks.
Can LLMs Write Correct and Efficient GPU Communication Code?
Winner 🏆 (Agent-only) MLSys 2026 - FlashInfer AI Kernel Generation Contest for the DeepSeek Sparse Attention (DSA) track with an average speedup of 34.93x
A public repo of an evaluations research project
RouterArena: An open framework for evaluating LLM routers with standardized datasets, metrics, an automated framework, and a live leaderboard.