Research Scientist (AI Safety)

TL;DR

Research Scientist (AI Safety): Developing post-training methods and interpretability techniques to make frontier AI systems safer and more transparent with an accent on model robustness and alignment properties. Focus on designing post-training pipelines, creating interpretability-informed evaluations, and translating research into actionable safety standards.

Location: San Francisco, CA; New York, NY; Seattle

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

Company

hirify.global is a leading data and evaluation partner for frontier AI companies, focusing on providing high-quality data and technology to develop reliable AI systems for critical decisions.

What you will do

• Design and run post-training pipelines to study how training choices affect model safety, robustness, and alignment.
• Develop interpretability-informed evaluations to reveal and mitigate unsafe, deceptive, or undesirable model behaviors.
• Collaborate with policymakers and engineers to translate research findings into safety standards and benchmarks.
• Tackle complex problems in agent robustness, AI control protocols, and AI risk evaluations.
• Publish findings to help governments and industry understand and mitigate AI risks.

Requirements

• Experience with post-training and RL techniques such as RLHF, DPO, GRPO.
• Track record of published research in machine learning, specifically in generative AI.
• At least three years of experience addressing sophisticated ML problems in research or product development.
• Strong written and verbal communication skills for operating in cross-functional teams.
• Deep commitment to promoting safe, secure, and trustworthy AI deployments.

Nice to have

• Experience with mechanistic interpretability, probing, or other model internal analysis techniques.
• Familiarity with red-teaming or adversarial evaluation of post-trained models.
• Experience studying failure modes like reward hacking, sycophancy, or alignment faking.

Culture & Benefits

• Comprehensive health, dental, and vision coverage.
• Equity compensation based on Board of Director approval.
• Retirement benefits and a learning and development stipend.
• Generous PTO and commuter stipend for eligible roles.

Hiring process

• Research interviews focused on practical ML prototyping and debugging.
• Assessment of research concepts and cultural alignment.
• No LeetCode-style questions used in the evaluation.