Performance & Systems Engineer, Codex
San Francisco • FullTime
Posted 4mo ago
Remote Work Policy
On-site
Employment Type
FullTime
Categories
Applied AI Engineer
About the job
The Codex team is building state-of-the-art AI systems for code generation and agentic reasoning, aiming to deploy these capabilities in products like ChatGPT and next-generation coding tools. As a Performance & Systems Engineer, you will focus on whole-system optimization across a complex stack including LLM inference, cloud orchestration, and agentic work management. Your role involves identifying and implementing high-leverage changes across infrastructure, modeling, and product layers to significantly improve the speed and cost-efficiency of Codex agents, directly impacting the experience of millions of users.
Responsibilities
- Identify and address inefficiencies across the Codex system stack, including agent behavior, LLM inference, and container orchestration.
- Develop tooling for measuring, profiling, and optimizing system performance at scale.
- Collaborate with researchers and engineers to implement changes that improve latency and reduce costs.
Requirements
- Experience operating across both ML systems and cloud infrastructure.
- Ability to dive into ambiguous problems and achieve clear results.
- Holistic approach to performance, balancing speed, cost, and user experience.
Benefits
- Relocation assistance for new employees.