Software Engineer, Safeguards Evals
Build evaluation infrastructure and datasets to measure how well AI agents detect misuse and policy violations. Design experiments, productionize evals into release pipelines, and improve safety investigation capabilities.
Key responsibilities
- Build and own the evaluation harness for an agentic investigation system — defining metrics, test cases and grading approaches for a complex long horizon agent
- Construct high-quality eval datasets representing real-world misuse across harm areas (e.g., cyber attacks, bio weapons, influence operations), drawing from real traffic patterns and synthetic generation
- Measure agent performance end-to-end (detection precision/recall, investigation quality, robustness) and drive hill-climbing on the hardest harm areas
- Analyze coverage to identify measurement gaps, and evolve evals so they remain unsaturated and high-signal as agent capabilities advance
- Productionize successful research into regression and release pipelines that run on every agent change, prompt update, and underlying model upgrade
- Build tooling that enables policy experts to author, run, and iterate on evaluations without engineering support
- Construct RL environments to improve Claude’s safety investigation capabilities
Minimum qualifications
- Proficiency in Python and comfort working across the stack
- Experience building and maintaining data pipelines
- Experience working with LLMs and a working understanding of their capabilities and failure modes — especially agentic systems with tool use and multi-step reasoning
- Strong data analysis skills — you can draw reliable insights from large datasets
- Ability to move fluidly between research prototyping and production-quality code
- Ability to translate ambiguous problems into concrete, testable experiments
Preferred qualifications
- 6+ years of industry software engineering experience
- Expertise in building or contributing to 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 on AI systems
- Experience with synthetic data generation or data augmentation
- Experience with distributed systems or large-scale data processing
- Experience with prompt engineering or building LLM-powered applications
Compensation and benefits
- Annual Salary: $320,000—$485,000 USD
- Location-based hybrid policy: Currently, we expect all staff to be in one of our offices at least 25% of the time. However, some roles may require more time in our offices.
- Visa sponsorship available
Researcher, Agent Post-Training, Personality
Researcher on the Agent Post-Training Personality team shaping how frontier agents communicate, collaborate, and build trust. Focuses on turning qualitative behavioral insights into evals, training data, reward signals, and model improvements.
Principal Engineer, AI Systems
Principal-level IC building and scaling production autonomous agents and agentic workflows across Block's ecosystem using frontier LLMs. Requires 15+ years experience shipping AI systems from zero to production scale.
Staff Software Engineer, Inference
Build and maintain distributed inference systems serving Claude to millions of users. Design intelligent routing, autoscaling, and high-performance infrastructure across diverse AI accelerators.
Researcher: Agent Post-Training, API & Power-Users
Improve agentic model capabilities for API and power users by designing experiments, building evals from real workflows, and driving post-training interventions from discovery through launch.