Low-Level AI Accelerator Runtime Engineer

AI Chopping Block; San Francisco, Northern (CA, KY); Hybrid; USD 180,000 - 240,000; Full time

Job summary

OpenAI is seeking a senior systems engineer to build the low-level device runtime for our custom silicon. You will design kernel-launch scheduling, memory management, and synchronization primitives, working closely with compiler, kernel, architecture, verification, and silicon teams.

You will use event-based cycle-accurate simulators to validate runtime behavior before silicon, diagnose concurrency and memory-ordering issues, and contribute to hardware-software interface improvements throughout

Qualifications

• Strong low-level systems programming experience in C, C++, Rust or comparable environments.
• Experience building runtimes, drivers, firmware, or adjacent systems infrastructure.
• Understanding concurrency, memory ordering, and hardware-software interfaces.

Responsibilities

• Design and implement the low-level device runtime for OpenAI custom silicon.
• Build kernel-launch scheduling, command submission, queueing, dependency tracking, and completion handling.

Skills

C/C++

Rust

System programming

Job description

OpenAI is seeking a senior systems engineer to build the low-level device runtime for our custom silicon. You will design kernel-launch scheduling, memory management, and synchronization primitives, working closely with compiler, kernel, architecture, verification, and silicon teams.

You will use event-based cycle-accurate simulators to validate runtime behavior before silicon, diagnose concurrency and memory-ordering issues, and contribute to hardware-software interface improvements throughout