SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
OpenAI's Frontier Assurance team is seeking a Researcher to build technical foundations for rigorous assessments of frontier AI risks and safeguards. This role combines research engineering, experimental design, and close technical collaboration to strengthen OpenAI's safety decisions through independent expertise and rigorous assessment.
You will design technical assessments with external evaluators and internal teams by identifying consequential safety claims to test, defining evidence and methods needed, and addressing coverage gaps. You'll work closely with external evaluators and embedded auditors, helping them understand systems being assessed and providing access to models, data, and technical information. Where direct access isn't possible, you'll conduct experiments and analyses on their behalf with methods they can meaningfully scrutinize.
Key responsibilities include:
- Designing assessments with external evaluators and internal teams to identify safety claims, define evidence requirements, and address gaps
- Supporting technical execution of third-party assessments and embedded audits on frontier risks
- Pairing with external researchers to provide data, documentation, and context for model and safeguard assessment
- Designing and implementing secure access to assessment environments across research, engineering, and security teams
- Conducting experiments and analyses when direct access is constrained, developing methods collaboratively and ensuring reproducibility
- Building tooling to automate auditing, assessment workflows, and internal assurance work
- Reproducing external findings, investigating discrepancies, and testing proposed mitigations
- Documenting methods, findings, and limitations clearly for evaluators and decision-makers
You should have strong software engineering skills and experience building research tools, evaluation infrastructure, or systems for analyzing complex data. Practical experience with modern machine learning systems, including designing experiments and running evaluations, is essential. You can navigate unfamiliar codebases and infrastructure, debug across system boundaries, and turn ambiguous research requests into reliable technical implementations. Sound judgment in experimental design and measurement—identifying confounders, limitations, and evidence gaps—is critical. You enjoy working closely with researchers, explaining technical systems, and supporting rigorous assessments that challenge assumptions. Experience with AI safety evaluations, adversarial testing, security research, or technical auditing is especially relevant.
Requirements:
- Strong software engineering skills and experience building research tools, evaluation infrastructure, or systems for analyzing complex data
- Practical experience with modern machine learning systems, including designing experiments, running evaluations, and interpreting results
- Ability to navigate unfamiliar codebases and infrastructure, debug across system boundaries, and implement reliable technical solutions
- Sound judgment in experimental design and measurement, including identifying confounders and evidence gaps
- Ability to work closely with researchers and help others investigate difficult questions
- Understanding of or strong interest in frontier AI risks, alignment, and safeguard evaluation
- Experience with AI safety evaluations, adversarial testing, security research, or technical auditing (especially relevant)
