Low-Level Runtime Engineer for AI Accelerator Silicon

OpenAI; United States; Remote; USD 160,000 - 220,000; Full time

Job summary

OpenAI is seeking a software-hardware integration engineer to build the low-level device runtime for our AI accelerator. You will bridge software and silicon to deliver efficient execution, memory management, and robust interfaces across runtimes and kernels.

You will design kernel-launch scheduling, memory lifetimes, synchronization primitives, and cycle-accurate simulators to validate performance before silicon is available and to tune behavior after deployment.

Responsibilities

• Design and implement the low-level device runtime for OpenAI custom silicon.
• Build kernel-launch scheduling, command submission, queueing, and completion handling.
• Manage device memory spaces, allocations, and virtual-to-physical mappings across concurrent workloads.
• Implement synchronization primitives, events, barriers, streams, and ordering guarantees.
• Define clean interfaces between the runtime, drivers, firmware, and higher-level execution systems.
• Use event-based, cycle-accurate simulators to develop, validate, and performance-tune runtime behavior.

Job description

OpenAI is seeking a software-hardware integration engineer to build the low-level device runtime for our AI accelerator. You will bridge software and silicon to deliver efficient execution, memory management, and robust interfaces across runtimes and kernels.

You will design kernel-launch scheduling, memory lifetimes, synchronization primitives, and cycle-accurate simulators to validate performance before silicon is available and to tune behavior after deployment.