Forward Deployed Engineer (FDE) - AI Infra - SF

About the team

OpenAI’s Forward Deployed Engineering team partners with customers to turn research breakthroughs into production systems. We operate at the intersection of customer delivery and core platform development.

About the role

Forward Deployed Engineers lead complex deployments of frontier models in production. You will embed with customers where model performance matters, delivery is urgent, and ambiguity is the default. You will use this to map their problems, structure delivery, and ship fast.

As a Forward Deployed Engineer for AI Infra, you will work directly with customer engineers, researchers and product engineers on optimization problems across semiconductors and AI infrastructure. You will write and debug software, run experiments against real workloads and hardware, and improve the tools and product used across engagements.

You will help define the problem and success criteria, build the software needed to test an approach, and stay with the work through validation and deployment. Work may include inference performance and chip-design optimization. Success means producing results that stand up to correctness and performance checks, and contributing improvements that other engagements can use.

Success in this role means owning the delivery state across workstreams. You will hold the bar on quality and pace and help OpenAI learn through execution.

This role is based in San Francisco. We use a hybrid work model of 3 days in the office per week. We offer relocation assistance. Travel up to 50% is required.

In this role, you will:

• Own technical delivery for agreed semiconductor and AI-infrastructure workstreams, from initial scoping and prototypes through validation, production deployment and adoption.

• Work closely with customer engineers to understand their workflows and constraints, define workloads, baselines and success criteria, and help what you build become part of how they work.

• Scope and sequence delivery, identify dependencies and blockers early, and make practical trade-offs between scope, speed and quality as the work develops.

• Personally build and debug production-quality software, including harnesses, integrations and product features that apply frontier models to problems in chip design, verification, inference and performance engineering.

• Design reproducible experiments, evaluations, graders and benchmarks to test approaches, validate correctness and performance, and establish whether improvements hold up in real engineering environments.

• Investigate model behavior and failure modes alongside researchers, contributing evaluation or training data and targeted experiments where they can improve the underlying capability.

• Contribute directly to core product development, working with product engineers to resolve tooling gaps, ship useful improvements and bring field learning into product decisions.

• Lead technical discussions and maintain alignment across customer, Product and Research teams. Build the domain and product fluency to explain capabilities, limitations and results, and keep ownership, risks and next steps clear.

• Turn successful approaches into reusable software, tools and playbooks, reducing the effort needed for future engagements and helping customer teams operate what has been built.

You might thrive in this role if you:

• Bring 5+ years of engineering or technical deployment experience that includes customer-facing work

• Have scoped and delivered complex systems in fast-moving or ambiguous environments

• Write and review production-grade software, and debug unfamiliar systems across code, integrations and runtime behavior.

• Have built or deployed systems powered by LLMs or generative models and understand how model behaviour affects product experience

• Simplify complexity and make fast, sound decisions under pressure

• Communicate clearly with engineers, product teams, and customer stakeho...