Kernel Engineer: GPU Performance & Inference

Acceler8 Talent; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Acceler8 Talent in San Francisco seeks a Member of Technical Staff focusing on Kernels & GPU Performance to push the limits of production AI inference across diverse accelerators.

You will implement low-level kernels, analyze memory hierarchies, and collaborate with compiler, ML systems, and runtime teams to optimize latency and throughput.

This on-site role offers a full-time path at a fast-growing AI infrastructure company.

Qualifications

• Bachelor’s degree in a relevant technical discipline or equivalent practical experience.
• Strong fundamentals in software engineering and performance.
• Experience with low-level execution, memory hierarchies, and scheduling.
• Ability to profile, debug, and optimize latency and throughput.
• Experience across multiple accelerator architectures.
• Familiarity with CUDA/Triton/CUTLASS is a plus.

Responsibilities

• Build and optimize kernels for production AI workloads.
• Improve latency, throughput, and hardware utilization.
• Develop execution strategies across multiple accelerator architectures.
• Optimize memory efficiency, scheduling behavior, and low-level execution characteristics.
• Analyze accelerator execution models and memory hierarchies to identify bottlenecks.
• Profile and validate performance across different hardware platforms.
• Partner with compiler, runtime, ML systems, and distributed-systems engineers on end-to-end performance optimization.
• Develop optimization approaches that account for architectural differences between accelerators.
• Help establish performance-engineering standards and best practices across the execution platform.

Skills

Software-engineering fundamentals

Performance-critical systems

Low-level execution

Memory hierarchy understanding

Profiling & debugging

Education

Bachelor’s degree in a relevant technical discipline

Tools

CUDA

Triton

CUTLASS

Accelerator programming models

GPU profiling tools

Job description

Acceler8 Talent in San Francisco seeks a Member of Technical Staff focusing on Kernels & GPU Performance to push the limits of production AI inference across diverse accelerators.

You will implement low-level kernels, analyze memory hierarchies, and collaborate with compiler, ML systems, and runtime teams to optimize latency and throughput.

This on-site role offers a full-time path at a fast-growing AI infrastructure company.