Loading activity…
@jiazhihao
Building @lithos_ai. Assistant Professor CS @CarnegieMellon. PhD CS @Stanford. Systems and ML.
AI
Its creators say SentryWard checks synchronization, data races and resource lifetimes in compiled NVIDIA PTX.
AI
The developer says megakernels and DSpark speculative decoding run Qwen3.8-27B at that peak, and shares the code.
AI
The instructors say they’ll cover knowledge curation, compiler analysis and feedback loops. They’ve also curated an online book on the topic.
