About this role
<p style="min-height:1.5em"><strong>About the Team</strong></p><p style="min-height:1.5em">The Agent Safety team works to ensure that increasingly capable AI agents act safely, exercise sound judgment, and remain aligned with user intent. Our mission is to reduce the probability of severe unintended outcomes from increasingly capable AI agents while preserving their ability to act effectively and autonomously.</p><p style="min-height:1.5em">Our work spans three areas:</p><ul style="min-height:1.5em"><li><p style="min-height:1.5em">Training: Create training methods, environments and data that teach agents to make better decisions in consequential situations. We turn real-world failures into training signals that prevent similar incidents, and identify precursor behaviors and mitigations to address emerging risks.</p></li><li><p style="min-height:1.5em">Measurements: Build evaluations and production metrics that identify emerging risks and measure whether our interventions work.</p></li><li><p style="min-height:1.5em"><strong>Oversight</strong>: Develop oversight and system mitigation mechanisms that reduce harmful actions while preserving useful agent autonomy (for example future versions of <a target="_blank" rel="noopener noreferrer nofollow" href="https://alignment.openai.com/auto-review/">auto-review</a>).</p></li></ul><p style="min-height:1.5em"></p><p style="min-height:1.5em"><strong>About the Role</strong></p><p style="min-height:1.5em">This role focuses on oversight and system-level mitigations that enable increasingly capable agents to operate safely and autonomously in real environments. We prioritize building oversight systems that are used in practice today, both internally and externally (see our recent work on <a target="_blank" rel="noopener noreferrer nofollow" href="https://alignment.openai.com/auto-review/"><u>action monitoring for codex</u></a> and former <a target="_blank" rel="noopener noreferrer nofollow" href="https://alignment.openai.com/scaling-code-verification/"><u>code review</u></a>). We also study longer-term questions about how increasingly capable agentis systems can be supervised, constrained, and corrected.</p><p style="min-height:1.5em">We’re looking for a safety&security minded researcher or engineer who can reason rigorously about security boundaries and agent behavior, then build and test practical mitigations. A background in AI control or security is welcome but not required.</p><p style="min-height:1.5em">This role is based in San Francisco, CA. We use a hybrid work model of 3 days in the office per week and offer relocation assistance to new employees.<br /></p><p style="min-height:1.5em"><strong>In this role, you will:</strong></p><ul style="min-height:1.5em"><li><p style="min-height:1.5em">Design, build, and evaluate system-level controls for agent actions like agent-based review. Plan how they fit in a broader system including sandboxing with process isolation and permission boundaries.</p></li><li><p style="min-height:1.5em">Work closely with a Codex harness engineering team to productionize the AI controls.</p></li><li><p style="min-height:1.5em">Red-team end-to-end agentic systems to measure whether controls prevent data exfiltration, unsafe tool use, and other harmful outcomes.</p></li><li><p style="min-height:1.5em">Improve the safety–productivity tradeoff by measuring and reducing missed harmful actions, unnecessary blocks, approval burden, and latency.</p></li></ul><p style="min-height:1.5em"></p><p style="min-height:1.5em"><strong>You might thrive in this role if you:</strong></p><ul style="min-height:1.5em"><li><p style="min-height:1.5em">Have strong systems or security instincts and can reason concretely about isolation boundaries, permissions, attack surfaces, and failure modes in complex systems.</p></li><li><p style="min-height:1.5em">Enjoy turning ambiguous safety questions into concrete threat models, reproducible experiments, and practical mitigations, and revising your approach based on evidence from deployment.</p></li><li><p style="min-height:1.5em">Can build robust experimental infrastructure and design evaluations that distinguish promising mitigations from brittle ones.</p></li><li><p style="min-height:1.5em">Are deeply interested in frontier AI alignment, safety and control.</p></li></ul><p style="min-height:1.5em"></p><p style="min-height:1.5em"><strong>About OpenAI</strong></p><p style="min-height:1.5em">OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity. We push the boundaries of the capabilities of AI systems and seek to safely deploy them to the world through our products. AI is an extremely powerful tool that must be created with safety and human needs at its core, and to achieve our mission, we must encompass and value the many different perspectives, voices, and experiences that form the full spectrum of humanity. </p><p style="min-height:1.5em">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. </p><p style="min-height:1.5em">For additional information, please see <a target="_blank" rel="noopener noreferrer" class="c-link c-link--underline" href="https://cdn.openai.com/policies/eeo-policy-statement.pdf"><u>OpenAI’s Affirmative Action and Equal Employment Opportunity Policy Statement</u></a>.</p><p style="min-height:1.5em">Background checks for applicants will be administered in accordance with applicable law, and qualified applicants with arrest or conviction records will be considered for employment consistent with those laws, including the San Francisco Fair Chance Ordinance, the Los Angeles County Fair Chance Ordinance for Employers, and the California Fair Chance Act, for US-based candidates. For unincorporated Los Angeles County workers: we reasonably believe that criminal history may have a direct, adverse and negative relationship with the following job duties, potentially resulting in the withdrawal of a conditional offer of employment: protect computer hardware entrusted to you from theft, loss or damage; return all computer hardware in your possession (including the data contained therein) upon termination of employment or end of assignment; and maintain the confidentiality of proprietary, confidential, and non-public information. In addition, job duties require access to secure and protected information technology systems and related data security obligations.</p><p style="min-height:1.5em">To notify OpenAI that you believe this job posting is non-compliant, please submit a report through <a target="_blank" rel="noopener noreferrer nofollow" href="https://form.asana.com/?d=57018692298241&k=5MqR40fZd7jlxVUh5J-UeA">this form</a>. No response will be provided to inquiries unrelated to job posting compliance.</p><p style="min-height:1.5em">We are committed to providing reasonable accommodations to applicants with disabilities, and requests can be made via this <a target="_blank" rel="noopener noreferrer" class="c-link c-link--underline" href="https://form.asana.com/?k=bQ7w9h3iexRlicUdWRiwvg&d=57018692298241"><u>link</u></a>.</p><p style="min-height:1.5em"><a target="_blank" rel="noopener noreferrer nofollow" class="c-link" href="https://cdn.openai.com/policies/global-employee-and-contractor-privacy-policy.pdf">OpenAI Global Applicant Privacy Policy</a></p><p style="min-height:1.5em">At OpenAI, we believe artificial intelligence has the potential to help people solve immense global challenges, and we want the upside of AI to be widely shared. Join us in shaping the future of technology.</p>