Research Engineer, Frontier Red Team
Description For Research Engineer, Frontier Red Team
Anthropic is seeking a Research Engineer for their Frontier Red Team to develop and implement "gold standard" evaluations for catastrophic risks in AI systems. This role is crucial for implementing the company's Responsible Scaling Policy (RSP) and ensuring the safe deployment of frontier AI models. The position involves creating evaluation systems for some of the most capable AI systems ever built, collaborating with experts across biosecurity, autonomous replication, cybersecurity, and national security domains.
The ideal candidate will combine strong engineering capabilities with a dedication to AI safety. You'll be responsible for building and scaling evaluation infrastructure that could become industry standards, while working in a collaborative environment with researchers and domain experts. The role requires both technical expertise and an understanding of the broader implications of AI safety.
Working at Anthropic means joining a cohesive team focused on high-impact AI research, treating it as an empirical science similar to physics and biology. The company values communication skills and collaboration, hosting frequent research discussions to ensure focus on the highest-impact work. Their research continues important directions including GPT-3, Circuit-Based Interpretability, and AI Safety.
The position offers competitive compensation (€225,000 - €270,000), comprehensive benefits, and a hybrid work environment requiring at least 25% office presence in San Francisco. Anthropic provides visa sponsorship and emphasizes diversity and inclusion, encouraging applications from candidates who might not meet every qualification but are passionate about AI safety and responsible development.
Responsibilities For Research Engineer, Frontier Red Team
- Design and implement robust evaluation infrastructure to measure model capabilities and risks across multiple domains
- Lead technical projects to build and scale evaluation systems
- Collaborate with domain experts to translate insights into concrete evaluation frameworks
- Build sandboxed testing environments and automated pipelines for continuous model assessment
- Work closely with researchers to rapidly prototype and iterate on new evaluation approaches
- Partner with cross-functional teams to advance Anthropic's safety mission
- Contribute to Capability Reports that inform critical deployment decisions
Requirements For Research Engineer, Frontier Red Team
- Experience leading and conducting fast, iterative experiments with frontier AI models
- Experience designing or implementing evaluations involving LLM sampling and prompting
- Strong software engineering skills with extensive Python experience
- Experience working with distributed systems
- Ability to write clean, well-structured code
- Strong interest in AI safety and responsible development
- Self-starter mentality and comfort in fast-paced environments
- Ability to balance urgency with careful implementation
Benefits For Research Engineer, Frontier Red Team
- Competitive compensation and benefits
- Optional equity donation matching
- Generous vacation and parental leave
- Flexible working hours
- Office space for collaboration
Jobs Related To Anthropic Research Engineer, Frontier Red Team
