Gimlet Labs
Member of Technical Staff - Kernels
$150K–$390K
San Francisco, CA · On-site · Staff · Full-time · Posted Mar 10, 2026
- Education
- Bachelor's degree in a relevant field, or an equivalent combination of education, training, and professional experience.
About the job
Builds and optimizes low-level kernels and execution primitives to improve production AI inference performance across diverse accelerator architectures.
What you'll do
- Build and optimize kernels for latency, throughput, and hardware utilization in production AI workloads.
- Improve memory efficiency, scheduling, and execution across the inference stack.
- Partner with compiler, runtime, and distributed systems engineers on end-to-end performance optimization.
What they ask for
- Strong software engineering fundamentals.
- Experience with performance-critical systems close to hardware.
- Comfort reasoning about low-level execution behavior, memory hierarchies, and performance tradeoffs.
- Bachelor's degree in a relevant field, or an equivalent combination of education, training, and professional experience.
Nice to have
- Experience with CUDA, Triton, CUTLASS, or other accelerator programming models.
- GPU execution model knowledge, memory optimization, profiling, or multi-GPU/distributed execution experience.
Pay
$150K – $390K (base salary)
Tools
CUDA, Triton, CUTLASS
AI focus
Optimizes kernels and accelerator execution for production AI inference workloads.
Model-written summary · the original posting is authoritative