OpenAI is hiring a Software Engineer for the Human Alignment Team within our Future of Computing Research group. You will build the infrastructure, data systems, and evaluation foundations critical for developing next-generation multimodal models. This role focuses on creating robust systems for product-grounded research, including data pipelines, human feedback tooling, and evaluation platforms.
What You'll Do
- Build evaluation and data foundations for next-generation personalized and multimodal AI systems.
- Partner closely with researchers to turn fuzzy behavioral questions into rigorous evals, datasets, rubrics, and scorecards.
- Design and implement human-data pipelines, grader systems, and experiment infrastructure for product-grounded research.
- Create evaluation frameworks for subjective, contextual, and long-horizon behaviors.
- Develop reproducible pipelines for collecting, processing, joining, and analyzing multimodal signals from real-world studies and product usage.
- Help define what should count as meaningful progress, and build the systems that let the team measure it with confidence.
- Work across research, safety, design, and engineering to ensure that what we optimize for is both technically sound and human-centered.
- Prototype quickly, iterate on measurement frameworks, and improve the team’s ability to debug, compare, and trust behavioral results.
- Shape the infrastructure and methodology that future OpenAI products will rely on for personalization, adaptation, and evaluation.
What We're Looking For
- Strong software engineering fundamentals and experience building data, backend, ML, or evaluation systems.
- Excellent research taste and strong judgment about what is worth measuring, how to measure it, and when a metric is misleading.
- Rigorous about data quality, reproducibility, metric design, and empirical correctness.
Nice to Have
- Experience with human-in-the-loop systems, annotation pipelines, experimentation platforms, or evaluation tooling.
- Enjoy working on ambiguous, early-stage problems where the hardest part is often defining the right evaluation rather than implementing the obvious one.
- Motivated by human-centered AI and excited by the challenge of measuring behaviors that are subtle, contextual, and difficult to benchmark.
- Want to help define the research infrastructure behind a new category of AI products.
Team & Environment
This role is part of the Future of Computing Research team, an applied research team within the Consumer Devices group. You will work closely across research, engineering, design, product, and safety.
Benefits & Compensation
- Relocation assistance to new employees.
Work Mode
This is a hybrid position based in San Francisco, CA.
OpenAI is dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity and must be created with safety and human needs at its core. We encompass and value the many different perspectives, voices, and experiences that form the full spectrum of humanity. We are an equal opportunity employer, and we do not discriminate on the basis of race, religion, color, national origin, sex, sexual orientation, age, veteran status, disability, genetic information, or other applicable legally protected characteristic.
