Member of Technical Staff

We’re looking for a Member of Technical Staff with 0-6 YOE to join our team building AI performance engineers — agents that automatically optimize GPU kernels and ML workloads to run as efficiently as possible on any hardware. You’ll work at the intersection of AI agent systems and low-level GPU engineering, helping expand support to new hardware architectures and fulfilling growing demand from chip companies and cloud providers. This is a high-ownership, zero-bureaucracy role where you’re in direct contact with customers from week one.

Responsibilities

• Build and improve the agent framework for GPU kernel optimization — multi-turn tool use, state management, reward signals

• Write and debug GPU kernels and low-level compiler code to understand and improve agent output

• Develop integrations with GPU profilers and compiler toolchains across CUDA, HIP, and other hardware targets

• Design architecture for remote GPU execution across cloud providers

• Work directly with chip company and cloud provider stakeholders (AMD, AWS, DigitalOcean, and others) as a forward-deployed engineer

• Drive long-term research bets and push the technical frontier for new hardware architectures