About this role
Invisible Technologies is building a reinforcement learning capability to measure frontier AI models. As a Research Engineer, you will design and implement evaluation methodologies, benchmarks, and RL environments, directly influencing how AI capabilities are assessed. This role requires you to translate research questions into production systems, writing the code that proves out your designs.
What we are looking for 6
- Design benchmarks and RL environments for AI model evaluation
- Originate evaluation methodology and translate into scoring frameworks
- Write and ship production code for evaluation systems
- Build and maintain data pipelines for evaluation runs
- Familiarity with reinforcement learning and frontier model evaluation
- Track record of turning research questions into working systems