Research Software Engineer (AI Safety)

TL;DR

Research Software Engineer (AI Safety): Designing and building secure infrastructure for sensitive model evaluations with an accent on sandboxed execution and secure data pipelines. Focus on implementing RBAC, defense-in-depth security, and tooling to measure model capabilities in high-consequence domains.

Location: On-site in New York, San Francisco, or London

Company

Research lab making intelligence open and accessible for everyone to use, customize, and build on.

What you will do

• Design and build secure, sandboxed infrastructure for running sensitive model evaluations, including CBRN and other dangerous-capability domains.
• Build controlled data pipelines and storage using least-privilege, RBAC, encryption, and audit logging.
• Partner with safety researchers to translate evaluation designs into reliable and scalable systems.
• Develop eval-orchestration tooling and harnesses for high-throughput evaluations in isolated environments.
• Implement guardrails and monitoring to ensure sensitive work remains appropriately siloed.
• Write production-quality Python for high-throughput data processing and evaluation systems.

Requirements

• Strong software engineering skills in Python with a track record of building scalable infrastructure.
• Experience building sandboxed, isolated, or security-sensitive execution environments (containers, VM isolation).
• Solid grounding in security engineering: RBAC, secrets management, encryption, and defense-in-depth.
• Experience building data pipelines for handling sensitive or restricted data.
• Ability to own ambiguous, cross-functional problems end-to-end.
• Must be based in or able to work on-site in New York, San Francisco, or London.

Nice to have

• Experience building evaluation, benchmarking, or experimentation infrastructure for ML systems.
• Experience working with LLMs, agents, or ML training/inference pipelines.
• Familiarity with dangerous-capability domains (CBRN, cyber) and information security.
• Familiarity with compliance frameworks for sensitive data handling.

Culture & Benefits

• Top-tier compensation with salary and stock options for all contributors.
• Comprehensive medical, dental, vision, and life insurance with an annual wellness allowance.
• Unlimited paid time off in the U.S. and 30 days in the U.K.
• 22 weeks paid parental leave for all new parents (including adoptive and surrogate).
• Daily lunch and dinner provided in the office.
• Visa sponsorship support for exceptional talent.