Staff+ Software Engineer, Safeguards Evals

Highlights

• Build eval methods that gate model training and launch decisions.
• Own evaluation harness for agentic system investigating misuse.
• Build datasets grounded in real harm like cyber and bio threats.

What they offer

• Competitive compensation and benefits
• Optional equity donation matching
• Generous vacation and parental leave
• Flexible working hours

Ideal candidate

Engineer-researcher hybrid who moves between experiments and production and cares deeply about AI safety. Comfortable with ambiguity and turning real-world misuse patterns into rigorous, trustworthy evaluations.

Requirements

• Experience building and maintaining data pipelines
• Experience working with LLMs and understanding of capabilities and failure modes, especially agentic systems
• Strong data analysis skills with ability to draw insights from large datasets
• Ability to move between research prototyping and production-quality code
• Ability to translate ambiguous problems into concrete, testable experiments
• Care deeply about AI safety and real-world impact
• 8+ years of industry software engineering or ML engineering experience
• Expertise in building LLM or agent evaluation frameworks, benchmarks, or automated grading systems
• Extensive experience in trust and safety, content moderation, or abuse detection systems
• Experience in red teaming, adversarial testing, or jailbreak research