Kernel Performance Engineer - Production AI Inference

Gimlet Labs, Inc.; San Francisco (CA); On-site; USD 180,000 - 270,000; Full time

Job summary

Gimlet Labs, Inc. is seeking a Member of Technical Staff to build and optimize low-level execution primitives for production AI inference across accelerators.

You will work near hardware across kernels, memory access, and profiling, collaborating with compiler, ML systems, and distributed engineers to push latency, throughput, and efficiency across diverse architectures.

This role offers the chance to influence how heterogeneous hardware is deployed in the next generation of AI infrastructure.

Qualifications

• Strong software engineering fundamentals.
• Experience with performance-critical systems near hardware.
• Understanding memory hierarchies and GPU execution models.
• Bachelor's degree or equivalent experience.

Responsibilities

• Build and optimize kernels to improve latency, throughput, and hardware utilization for production AI workloads.
• Develop execution strategies that unlock performance across established and emerging accelerator architectures.
• Improve memory access patterns, scheduling, and execution characteristics across the inference stack.
• Collaborate with compiler, runtime, and distributed systems engineers to ensure end-to-end performance optimization.
• Influence how heterogeneous hardware is deployed and utilized within the next generation of AI infrastructure.
• Help establish performance engineering standards that shape Gimlet's execution platform.

Skills

Strong software

Performance systems

Low-level execution

Memory hierarchies

Profiling expertise

Education

Bachelor's degree in a relevant field

Tools

CUDA

Triton

CUTLASS

GPU profiling tools

Job description

Gimlet Labs, Inc. is seeking a Member of Technical Staff to build and optimize low-level execution primitives for production AI inference across accelerators.

You will work near hardware across kernels, memory access, and profiling, collaborating with compiler, ML systems, and distributed engineers to push latency, throughput, and efficiency across diverse architectures.

This role offers the chance to influence how heterogeneous hardware is deployed in the next generation of AI infrastructure.