Mountain View, CAHybrid$200,000 - $350,0001 months ago
full-timeseniorllamamistralcustom
About the Role
Build the software that makes Groq's LPU the fastest inference chip. Work on the compiler that translates models to run on custom silicon.
What you'll do:
- Develop compiler optimizations for LLM inference
- Implement model parallelism strategies
- Optimize memory bandwidth utilization
- Work with hardware teams on next-gen architecture
Required Skills
PythonC++CUDALLVM
About Groq
AI inference company building the fastest LPU chips for real-time AI.