Research Engineer - AI Performance & Kernel Optimization

Highlights

• Optimize large-scale LLM training and inference for frontier-scale AI systems.
• Hunt bottlenecks in memory, communication, and compute to boost throughput.
• Turn low-level systems wins into real-world training and inference gains.

What they offer

• Comprehensive medical, dental, vision, and FSA plans
• 401(k) plan
• Relocation and immigration support on a case-by-case basis
• In-office snacks and meals provided
• Unlimited PTO and company holidays

Ideal candidate

Deep systems thinker who loves chasing the last 10-20% of performance and reasoning about hardware-software tradeoffs. Thrives collaborating across research and infra to translate low-level wins into frontier-scale impact.

Requirements

• Strong engineering aptitude for building reliable, high-performance systems
• Excellent low-level performance intuition for hardware-software interactions
• Eager to rapidly learn new systems, tools, and hardware environments
• Excellent communication and collaboration skills across research and engineering
• Enjoy hunting down the last 10-20% of performance
• Experience writing highly performant GPU kernels
• Experience optimizing ML workloads for large-scale training and inference
• Understanding of distributed training and parallelism schemes
• Experience using profiling and debugging tools for performance improvements