Research Scientist (AI Controls and Monitoring)

TL;DR

Research Scientist (AI Controls and Monitoring): Designing methods, systems, and experiments to ensure advanced AI models remain aligned with intended goals with an accent on monitoring techniques, observability, and layered control protocols. Focus on building prototype systems, developing red-team simulations, and establishing benchmarks for AI monitoring.

Location: Must be based in San Francisco, New York, or Seattle

Salary: $216,000 - $270,000 USD

Company

A leading data and evaluation partner for frontier AI companies, providing technologies to help enterprises and governments build and oversee reliable AI applications.

What you will do

• Develop real-time monitoring and observability methods to track AI behavior and flag deviations or anomalous outputs.
• Research layered control mechanisms, including fail-safes and oversight protocols to redirect AI systems when risks are detected.
• Design red-team simulations to probe weaknesses in oversight and control mechanisms and build corresponding mitigations.
• Collaborate with policymakers, engineers, and researchers to establish global standards and benchmarks for AI monitoring.

Requirements

• Must be based in San Francisco, New York, or Seattle.
• Proven track record of published research in machine learning, particularly in generative AI.
• At least three years of experience addressing sophisticated ML problems in research or product development.
• Practical experience designing control and monitoring experiments and building working prototypes.
• Strong written and verbal communication skills for cross-functional collaboration.

Nice to have

• Experience with runtime monitoring, anomaly detection, or observability for ML systems.
• Familiarity with AI control or alignment research (e.g., scalable oversight, interpretability, debate).
• Experience with post-training and RL techniques such as RLHF, DPO, and GRPO.

Culture & Benefits

• Comprehensive health, dental, and vision coverage.
• Retirement benefits and a dedicated learning and development stipend.
• Generous PTO and commuter stipends for eligible roles.
• Equity-based compensation granted based on performance and approval.

Hiring process

• Interviews assessing practical ML prototyping, debugging, and grasp of research concepts.
• Cultural fit discussions to ensure alignment with organizational values.
• No LeetCode-style technical questions.