TPU Kernel Engineer

Highlights

• Design and optimize TPU kernels for research, training, and inference systems.
• Build low-latency, high-throughput sampling for large language models.
• Debug kernel performance at assembly level and model system performance.

What they offer

• Competitive compensation and benefits
• Optional equity donation matching
• Generous vacation and parental leave
• Flexible working hours
• Collaborative office space

Ideal candidate

Results-oriented systems engineer who loves low-level optimization and pair programming, eager to learn ML research and driven by building safe, beneficial AI with societal impact in mind.

Requirements

• Significant experience optimizing ML systems for TPUs, GPUs, or other accelerators
• Track record of solving large-scale systems problems and low-level optimization
• Bachelor's degree or equivalent combination of education, training and experience
• Background in a field relevant to the role
• Experience with high performance, large-scale ML systems
• Experience designing and implementing kernels for TPUs or other ML accelerators
• Deep understanding of accelerators, e.g. background in computer architecture
• Knowledge of ML framework internals
• Experience with language modeling with transformers