Offensive Cybersecurity Engineers: Build RL Environments for AI Evaluation
- Location
- United States
- Work arrangement
- Remote
- Employment type
- Contract
- Category
- Information Technology
- Industry
- Information Technology
- Salary
- $150 / hour
- Openings
- 1
- Apply by
- 9 Dec 2026
About the role
Terac is hiring offensive cybersecurity engineers to build reinforcement-learning environments that test and evaluate frontier AI models. This is a long-term, fully remote project paid by the hour: you will design and implement complex cybersecurity benchmarks, create robust RL environments for offensive security scenarios, and harden them so they stand up to advanced AI interactions. Your work will help researchers and AI labs measure how capable — and how safe — their models really are. You will collaborate with Terac's team over the long term, iterating on security scenarios and refining the evaluation criteria that decide whether a model passes. The role suits an experienced penetration tester or security researcher who enjoys hands-on technical work, cares about AI safety, and wants flexible remote work on a meaningful project.
Responsibilities
- Design and build offensive cybersecurity reinforcement-learning environments. - Harden existing benchmarks so they withstand advanced AI interactions. - Develop rigorous evaluation criteria for frontier AI models. - Create robust RL environments for offensive security scenarios. - Collaborate with Terac's team over the long term to iterate on security scenarios. - Document environments and evaluation methods so results are repeatable.
Requirements
- Professional experience in offensive cybersecurity or penetration testing. - Strong background in building and hardening technical environments. - Comfortable working on long-term, hourly, fully remote projects. - A genuine interest in AI safety and model evaluation. - Able to work independently and communicate clearly with a distributed team.
Preferred qualifications
- Experience with reinforcement learning or AI evaluation. - Background as a security researcher. - Experience designing benchmarks or evaluation frameworks.
Benefits
- Fully remote, long-term hourly work with flexible hours. - $150 per hour compensation. - Work on frontier AI evaluation with researchers and AI labs.
Skills
Nice to have:
How to apply
- Select Apply now.
- You are taken to Terac’s own application page to send your application.
Applying is free. We never ask candidates for money. The exact steps can differ by job and employer.