About the job
OpenAI is hiring a Tokens-as-a-Service (TaaS) Software Engineer in San Francisco. This position focuses on building and maintaining systems that transform large-scale infrastructure capacity into reliable, measurable token throughput for demanding workloads. The work combines hands-on engineering with a strong focus on operational reliability and performance.
The engineer will work on performance benchmarking, tokenomics, model porting, infrastructure integration, systems tooling, and operational monitoring. The role bridges partner and in-house compute environments with OpenAI’s infrastructure, supporting onboarding, measurement, monitoring, and optimization of GPU resources for real-world workloads.
What you will do
- Design and develop systems and tools to measure, monitor, and improve token throughput across a range of compute environments.
- Support benchmarking, tokenomics analysis, and model porting for different infrastructure setups.
- Create integration tools to connect external or partner infrastructure with OpenAI’s compute, observability, and workload management systems.
- Set up and track operational metrics such as billing, usage, SLAs, utilization, reliability, and throughput.
- Identify and resolve bottlenecks in hardware, networking, software, or workload enablement that impact token generation.
- Work closely with compute, infrastructure, networking, finance, and operations teams to turn raw capacity into usable workload-serving power.
- Build dashboards, automation, and reporting tools that provide clear visibility into TaaS capacity, efficiency, and business results.
Requirements
- Strong background in software engineering, with direct experience building systems, tools, automation, or infrastructure platforms.
- Experience working with compute infrastructure, distributed systems, performance engineering, or operational production environments.
- Analytical skills for assessing token throughput, utilization, benchmarking, infrastructure efficiency, and workload performance.
- Ability to integrate external systems or partner environments into internal infrastructure stacks.
- Excellent debugging and analytical abilities across hardware, networking, software, and operational areas.
Preferred qualifications
- Experience with cloud platforms and services.
- Knowledge of machine learning frameworks and token management systems.

