Sign up to access all features of our service.
  • Job search
  • Favorites
  • Create a CV
    New
  • Salaries
  • Subscriptions

Researcher, Agent Safety, Training and Evaluations

Full-time

OpenAI

About the Team

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.

Our work spans three areas:

  • 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.

  • Measurements: Build evaluations and production metrics that identify emerging risks and measure whether our interventions work.

  • Oversight: Develop oversight and system mitigation mechanisms that reduce harmful actions while preserving useful autonomy (for example future versions of auto-review ).

About the Role
We’re looking for strong executors with excellent judgment, comfort with ambiguity, and an understanding of frontier model research. You don’t need prior safety or alignment experience, we also welcome people that recently realized that alignment and safety is a critical area to contribute to.

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.

In this role, you will:

  • Train and evaluate frontier models to reduce harmful or misaligned agent actions, forming clear hypotheses and executing independently through ambiguity.

  • Mine incidents and build scalable measurement, data-processing, and evaluation systems that turn real failures into repeatable safety signals.

  • Collaborate closely with post-training, capabilities, oversight, and pre-training partners to ship research-backed mitigations into large-scale training and agent systems.

You might thrive in this role if you:

  • Have demonstrated strength in research engineering, ML engineering, quantitative research, or applied model research, with the ability to own ambiguous projects end to end.

  • Bring excellent technical execution across experimentation, data, evaluation, and/or infrastructure, plus strong intuition for modern frontier-model research.

  • Are motivated by agent safety and eager to work on urgent, practical problems even if your prior work was outside safety or alignment.

About OpenAI

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. 

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.

For additional information, please see OpenAI’s Affirmative Action and Equal Employment Opportunity Policy Statement .

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.

To notify OpenAI that you believe this job posting is non-compliant, please submit a report through this form . No response will be provided to inquiries unrelated to job posting compliance.

We are committed to providing reasonable accommodations to applicants with disabilities, and requests can be made via this link .

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.

Vacancy posted 5 days ago
Similar jobs that could be interesting for youBased on the Researcher, Agent Safety, Training and Evaluations in San Francisco, CA vacancy
  •  ...About the Team The Agent Safety team works to ensure that increasingly...  ...work spans three areas: Training: Create training methods,...  ...risks. Measurements: Build evaluations and production metrics that...  ...for a safety&security minded researcher or engineer who can reason rigorously... 
    Training
    Full time
    Work at office
    Relocation package

    openai

    San Francisco, CA
    5 days ago
  •  ...work gets done, but bringing agents into a real organization remains...  .... Summary This is a research role for someone who wants to...  ...deployment, develop ways to evaluate and improve multiple agents operating...  .... The company is not training models today; its current research... 
    Training

    Greylock Partners

    San Francisco, CA
    3 days ago
  • $380k

    About the TeamSafety Research at OpenAI is responsible for ensuring...  ...in the real world. The Safety Training team develops the research and...  ...training, reinforcement learning, evaluation, and deployment feedback.Our...  ..., multimodal risks, agent behavior, and domain-specific... 
    Training
    Work at office
    Local area
    Flexible hours

    OpenAI

    San Francisco, CA
    1 day ago
  • $256.9k - $333.4k

     ...Intelligence and Security Research team to design the...  ...foundation where autonomous AI agents, enterprise software,...  ...& Guardrails: Design, evaluate, and deploy automated...  ...red-teaming and safety benchmarking across...  ...certifications, and/or training. The full salary range... 
    Training
    Full time
    Temporary work
    Work experience placement
    Local area
    Flexible hours

    CISCO Systems

    San Francisco, CA
    8 hours ago
  • Alterra Labs in the United States is seeking an ML Researcher specializing in post-training infrastructure to advance safety and alignment in healthcare and life sciences. You will design and evaluate post-training solutions and contribute to building simulation environments... 
    Training

    Alterra Labs

    San Francisco, CA
    2 days ago
  •  ...that assist humans to agents that can plan, execute...  ...measures, RSI-relevant training interventions, and...  ...Design experiments and evaluations to understand the extent...  ...misaligned, or their safety-relevant capabilities...  ...rigorous hypothesis-driven research and turning our... 
    Training

    OpenAI, Inc.

    San Francisco, CA
    1 day ago
  • TypeSafe AI is seeking a safety researcher to examine model behavior and help evaluate safety concerns within the team. You’ll build evals, document findings, and contribute to responsible research practices. During the six-week period, you’ll set up access, meet the team... 

    TypeSafe AI

    San Francisco, CA
    1 day ago
  • ML Researcher, Reinforcement Learning & Agents Endo Health San Francisco ·⏰ Full-time "Build the RL environment...  ..., reward design, post-training, and reproducible experiments. •...  ...environment, objectives, training, and evaluation. ✦ Help create the experience models... 
    Training
    Full time
    Relocation package

    Endo Health

    San Francisco, CA
    7 days ago
  • Job Title: ML Researcher, Post-Training Infrastructure About Alterra Labs: Alterra Labs is a frontier research lab for AI safety and alignment in healthcare and life sciences. We are looking for ML researchers with expertise in post-training to join our team. We build... 
    Training
    Full time

    Alterra Labs

    San Francisco, CA
    2 days ago
  • ### About the Team The [Safety Systems team]( is responsible for...  ...transparency. The Model Safety Research team aims to fundamentally...  ...topics such as RLHF, adversarial training, robustness, and more....  ...safety standards. Actively evaluate and understand the safety of... 
    Training
    Full time

    OpenAI

    San Francisco, CA
    5 days ago
  •  ...and respond appropriately. The Chat and Multimodal Safety team is responsible for ensuring that OpenAI’s increasingly...  ...safely across these experiences. We develop the research, training methods, and evaluations needed to make these experiences safe. Our work sits at... 
    Training
    Full time
    Work at office
    Relocation package

    openai

    San Francisco, CA
    5 days ago
  •  ...[measure monitorability]( which training mechanisms affect monitorability,...  ...monitoring can be a practical additional safety mechanism, and today our...  ...intersects with model training, alignment evaluations, monitoring, and frontier-risk research.We care most about monitorability... 
    Training
    Full time
    Work at office
    Relocation package

    OpenAI

    San Francisco, CA
    5 days ago
  • $169k - $250k

     ...collaboration, join us! We’re hiring a Researcher, AI Evaluations to join AI Insights within Figma’s...  .... You’ll strengthen the team’s model, agent system, and evaluation work through hands...  ...metrics for AI-powered features Advanced training in a quantitative or behavioral... 
    Training
    Minimum wage
    Full time
    Local area
    Remote work
    Flexible hours

    Figma

    San Francisco, CA
    5 days ago
  • $30 - $60 per hour

     ...Pay: $30 - $60/hour Required Skills Research & source triangulation Attention to detail Written...  ...is a data lab that helps AI labs & Enterprises train foundational models and agents. We provide frontier evaluations and reinforcement learning environments to... 
    Training
    Part time
    For contractors
    Remote work
    Worldwide

    micro1

    San Francisco, CA
    20 days ago
  • $23 - $23.5 per hour

     ...Warehouse AgentDo you enjoy working in a fast-paced, safety-obsessed aviation environment? As a Warehouse Agent, you will be essential to increase operational...  ...operating in a safe and responsible manner.Complete all training when required by company, airport governing... 
    Training
    Full time
    Part time
    Work experience placement
    Work at office
    Immediate start
    Flexible hours
    Night shift

    AGI

    San Francisco, CA
    4 days ago
  •  ...On-site We are seeking an AI Evaluation Engineer – Reinforcement Learning & Agents to build the environments, evaluation...  ...infrastructure used to train and assess long-horizon enterprise...  ...will work on the engineering and research problems behind realistic agent environments... 
    Training

    MaxIT Consulting - Max Corporate Group

    San Francisco, CA
    10 days ago
  •  ...dedicated to improving patient safety in hospitals through advanced...  ...member of the clinical AI research team at Pharos. As a Physician...  ...datasets, developing and evaluating AI models, validating outputs...  ...create annotation schemas, curate training datasets, and ensure data... 
    Training
    Full time
    Work at office

    Pharos

    San Francisco, CA
    5 days ago
  •  ...Service (USSS) as a Special Agent offers a unique and rewarding...  ...other protectees, ensuring their safety in every environment....  ...complete an intensive 13-week training at the Federal Law Enforcement...  ...period during which we will evaluate your fitness and whether your... 
    Training
    Overseas
    Trial period

    The United States Secret Service

    San Francisco, CA
    22 days ago
  •  ...Proximal is building the research systems needed to identify what...  ...labs to provide the data and evaluations behind their most capable models...  ...early team has built coding agents and RL infrastructure at companies...  ...weaknesses and curate post-training data to fix these weaknesses.... 
    Training

    Proximal LLC

    San Francisco, CA
    2 days ago
  •  ...Night Audit Agent Palisociety is a society of unique hotels and residences Palihouse...  ...employees our greatest asset: structured training from day one, consistent feedback, and a...  ...Engineering, and conduct walkthroughs for safety, security, and cleanliness. Reconcile... 
    Training
    Flexible hours
    Shift work
    Night shift
    Day shift

    Hotel Grand Stark

    San Francisco, CA
    3 days ago
  • $47.25 per hour

     ...professionals passionate about security and safety. Join our innovative team committed to...  ...key roles, such as executive protection agents, intelligence analysts, armed security operatives...  ...required. Previous medical training is highly desirable. Licensing requirements... 
    Training
    Hourly pay
    Full time
    Work at office
    Local area
    Flexible hours
    Shift work
    Night shift
    Weekend work

    Allied Universal® Enhanced Protection Services

    San Francisco, CA
    2 days ago
  • $150k - $250k

     ...goods, and global social organizations.We research and deploy technologies that power AI-...  ...builds architectures that continuously evaluate and enhance their own performance. Researchers...  ...systems using models rather than training or fine-tuning them. Ideal candidates have... 
    Training
    Work at office
    3 days per week

    Distyl AI

    San Francisco, CA
    1 day ago
  • $197.3k - $313.7k

     ...#1 AI CRM, where humans with agents drive customer success together...  ...is seeking a Principal Researcher to help us understand how AI...  ...work practices with practical evaluation of product experiences, translating...  ..., promotion, benefits, training, assessment of job performance... 
    Training
    Full time

    Salesforce

    San Francisco, CA
    1 day ago
  • $216.3k - $280.8k

     ...we are leading frontier AI research across Cisco. Our mission is...  ...systems, scalable training algorithms, evaluation science, inference optimization...  ...agentic AI workflows, multi-agent frameworks, and autonomous...  ...security-focused use cases, safety, and robust machine learning... 
    Training
    Full time
    Temporary work
    Local area
    Flexible hours

    CISCO Systems

    San Francisco, CA
    4 days ago
  • $148.5k - $223.9k

     ...#1 AI CRM, where humans with agents drive customer success together...  ...a Senior Offensive Security Researcher with an attacker mindset to...  ...help our recruiters assess and evaluate candidates’ resumes and...  ...compensation, promotion, benefits, training, assessment of job... 
    Training
    Full time

    Salesforce

    San Francisco, CA
    3 days ago
  • $75 per hour

     ...professionals passionate about security and safety. Join our innovative team committed to...  ...key roles, such as executive protection agents, intelligence analysts, armed security operatives...  .... Participate in ongoing operational training, professional development, and program... 
    Training
    Hourly pay
    Full time
    Work at office
    Local area
    Relocation
    Monday to Friday
    Flexible hours
    Shift work
    Night shift
    Weekday work

    Allied Universal® Enhanced Protection Services

    San Francisco, CA
    15 days ago
  • $216.3k - $280.8k

     ...considered.Meet the TeamThe Splunk Agent Resilience team is defining...  ...scalable, cost-effective evaluation and guardrails that ensure AI...  ...will independently own applied research problems that help...  ...dataset preparation through model training, evaluation, and implementation... 
    Training
    Full time
    Temporary work
    Local area
    Remote work
    Flexible hours

    CISCO Systems

    San Francisco, CA
    8 hours ago
  • $184k - $238.8k

     ...AI, we are leading frontier AI research across Cisco. Our mission is...  ...new model architectures, pre-training and post-training methods, inference optimization, evaluation techniques, and data pipelines...  ...automated research workflows, where agents build on previous results to... 
    Training
    Full time
    Temporary work
    Local area
    Flexible hours

    CISCO Systems

    San Francisco, CA
    2 days ago
  • The company is a research-led startup building the evaluations and training environments that leading AI labs use to test and improve their models. Since raising...  ...loops into how the company works Building agents that can analyse thousands of environments and millions... 
    Training

    Velocity Tech

    San Francisco, CA
    1 day ago
  • $25 - $45 per hour

     ...Skills ChatGPT Muse AI agents Attention to Detail...  ...that helps AI labs & Enterprises train foundational models and agents. We provide frontier evaluations and reinforcement learning environments...  ...-quality training data. Our research spans high-value, specialized... 
    Training
    Part time
    For contractors
    Work at office
    Weekend work

    micro1

    San Francisco, CA
    14 days ago

Do you want to receive more vacancies?

Subscribe and receive similar vacancies to Researcher, Agent Safety, Training and Evaluations. Be the first to apply!