We are seeking a Compute Optimization Researcher/Engineer to build the systems that maximize the value of OpenAI's global compute capacity.
In this role, you will work on high-impact optimization problems spanning capacity allocation, demand forecasting, cluster planning, workload placement, and infrastructure utilization. You will combine mathematical modeling, software systems, and cross-functional execution to improve how compute is planned and consumed across GPU clusters, networking, storage, and data center environments.
This role is ideal for candidates with backgrounds in operations research, optimization, applied math, infrastructure systems, or large-scale capacity planning.
Key responsibilities include:
- Building optimization models for compute allocation, workload scheduling, and cluster utilization.
- Developing planning systems that balance supply, demand, cost, latency, and reliability constraints.
- Creating forecasting frameworks for GPU demand, infrastructure growth, and capacity needs.
- Designing decision tools for allocating compute across internal teams, products, and strategic priorities.
- Partnering with architecture, infrastructure engineering, finance, and operations teams to translate business needs into mathematical models.
- Integrating multiple operational data sources into planning systems and optimization workflows.
- Improving utilization of GPUs, networking, power, cooling, and storage infrastructure.
- Analyzing tradeoffs across first-party data centers, cloud providers, and hybrid environments.
- Building dashboards, metrics, and operational tooling for capacity decision-making.
- Leading ambiguous, cross-functional initiatives that improve infrastructure efficiency at scale.
- Presenting recommendations clearly to technical leaders and executives.
Requirements include:
- A doctorate degree in Computer Science, Engineering, Mathematics, Operations Research, Economics, or related field.
- 5+ years of experience in optimization, planning, infrastructure analytics, or systems engineering.
- Strong experience with linear programming, mixed-integer optimization, convex optimization, simulation, or forecasting methods.
- Proficiency in Python and data tooling (SQL, Pandas, Spark, etc.).
- Experience translating real-world business constraints into scalable optimization systems.
- Strong analytical problem-solving skills with comfort operating in ambiguous environments.
- Ability to influence cross-functional stakeholders without formal authority.
- Excellent communication skills with both technical and non-technical audiences.
Preferred qualifications include experience with large-scale infrastructure, cloud capacity planning, or data center operations, familiarity with tools such as Gurobi, CPLEX, CVXPY, Pyomo, or similar solvers, and experience optimizing GPU fleets, networking systems, or distributed compute environments.
XML job scraping automation by YubHub
