Role Overview

Join OpenAI's Alignment team as a Researcher, Alignment in San Francisco, California, and become a pivotal force in ensuring AI systems are safe, trustworthy, and aligned with human values. This senior-level role is perfect for PhD-level experts in AI safety, machine learning, and cognitive science who thrive in fast-paced, collaborative environments. At OpenAI, you'll tackle the most pressing challenges in AI alignment, developing methodologies that allow AI to robustly follow human intent—even in adversarial or high-stakes scenarios.

The Alignment team focuses on two core pillars: (1) scaling alignment techniques alongside growing AI capabilities, and (2) centering humans through intuitive interfaces for intent expression and oversight. As capabilities advance, your work will ensure our models remain reliable in complex real-world deployments. This hybrid position (3 days/week in-office) offers relocation assistance and positions you at the forefront of AI research that shapes humanity's future.

Ideal candidates are team players with strong engineering skills in PyTorch, experience in scalable oversight, and a passion for trustworthy AI. If you're ready to design experiments measuring subjective alignment risks, build robustness tools, and pioneer human-AI paradigms, this Researcher, Alignment job at OpenAI is your chance to make history.

Key Responsibilities

As a Researcher, Alignment at OpenAI, your contributions will directly impact AI safety. Here's what you'll do daily:

• Develop and evaluate alignment capabilities for subjective, context-dependent challenges hard to quantify.
• Design rigorous evaluations to measure AI risks and alignment with human values.
• Build specialized tools to test model robustness across diverse scenarios.
• Engineer experiments exploring alignment scaling with compute, data volume, context/action lengths, and adversary resources.
• Innovate Human-AI interaction paradigms and scalable oversight methods for complex supervision.
• Train models to calibrate accurately on correctness and risk predictions.
• Pioneer novel AI-assisted approaches in alignment research.
• Implement scalable solutions integrating human oversight into advanced AI decisions.
• Analyze how alignment holds up as AI capabilities surge.
• Prototype interfaces enabling humans to express intent effectively.
• Quantify deployment risks in high-stakes environments.
• Collaborate cross-functionally to iterate on cutting-edge methodologies.
• Conduct adversarial testing to ensure model reliability.

These responsibilities demand versatility, from hands-on coding to theoretical research, in a dynamic San Francisco-based team.

Qualifications

To excel as a Researcher, Alignment at OpenAI, bring these qualifications:

• PhD or equivalent in computer science, computational science, data science, cognitive science, or allied fields.
• Expert engineering in large-scale ML systems, especially PyTorch optimization.
• Deep knowledge of alignment algorithms, techniques, and underlying science.
• Proficiency in data viz/collection tools (TypeScript, Python).
• Thrives in fast-paced, collaborative research settings.
• Passion for safe, reliable AI in high-stakes contexts.
• Experience designing scalable oversight and HAI paradigms.
• Track record in risk evaluation and model calibration.
• Strong experiment design for scaling laws in AI.
• Team-oriented mindset for diverse tasks.
• Familiarity with adversarial robustness testing.

Senior-level experience preferred; OpenAI seeks innovators ready to redefine AI safety.

Salary & Benefits

OpenAI offers competitive compensation for Researcher, Alignment roles, estimated at $250,000–$450,000 USD yearly (including base, bonus, equity), reflecting San Francisco market rates for top AI talent. Total comp varies by experience.

Benefits include:

• Equity in a leading AI firm.
• Full health/dental/vision coverage.
• Hybrid SF model + relocation aid.
• Unlimited PTO and flexible policy.
• Generous parental leave.
• 401(k) matching.
• Professional growth stipend.
• Wellness programs, gym access.
• Catered meals and events.
• Commuter subsidies.
• Stock purchase options.
• Impact-driven culture.

Why Join OpenAI?

OpenAI is the vanguard of AGI development, committed to benefiting humanity. Our Alignment team leads in solving AI's toughest safety puzzles. Work with world-class researchers in San Francisco, influencing models deployed globally. Enjoy hybrid flexibility, top-tier perks, and the mission to make AI trustworthy. This Researcher, Alignment role offers unparalleled impact—join us to align the future.

Candidates rave about OpenAI's innovative culture, resources, and real-world difference. With rapid growth, your work scales instantly.

How to Apply

Ready for Researcher, Alignment at OpenAI? Submit your resume, PhD details, and alignment research portfolio via our careers page. Highlight PyTorch projects, scaling experiments, and safety contributions. Interviews include technical deep-dives and team fits. OpenAI is equal opportunity—diversity drives innovation. Apply now for San Francisco hybrid role!