about 6 hours ago
Base Salary
$182k - $242k/yr
Responsibilities
- Author, profile, and optimize CUDA kernels for LLM inference.
- Optimize GPU performance by tuning occupancy, memory coalescing, and data movement.
- Prototype and ship kernels using kernel-authoring DSLs and compilers.
- Build reproducible microbenchmarks and validate performance gains.
- Implement and maintain benchmarking workflows for MLPerf runs.
- Lead design reviews and decompose multi-service work into milestones.
- Mentor junior engineers and elevate coding/testing standards.
Requirements
- 5+ years of experience in high-performance computing or GPU software.
- Hands-on experience with CUDA programming and memory model.
- Deep understanding of GPU architecture and performance optimization.
- Strong coding skills in C++ and Python.
- Familiarity with model-serving stacks like vLLM and TensorRT-LLM.
- Excellent communication skills for cross-functional collaboration.
Benefits
- 100% paid medical, dental, and vision insurance.
- Company-paid life insurance and disability insurance.
- Flexible Spending Account and Health Savings Account.
- Tuition reimbursement and employee stock purchase program.
- Mental wellness benefits and family-forming support.
- Flexible PTO and catered lunch in office locations.