RL Environment Engineers: Building Tool-Based Environments

We are running a paid research initiative focused on advancing reinforcement learning applications in knowledge work. Our goal is to create robust, tool-based environments that accurately simulate professional workflows. Your engineering expertise will directly shape the architecture and functionality of these new tool gyms.

In this role, you will collaborate remotely to design and implement interactive RL environments. You will map out knowledge-work tools and build the corresponding gym interfaces for agents to interact with. The work requires writing clean Python code, testing environment dynamics, and refining the state and action spaces based on system performance. Expect to dedicate approximately 20 hours per week to these core development tasks.

We are looking for experienced reinforcement learning engineers with a strong background in environment design and simulation. Professionals who have previously built custom gym interfaces or similar tool-based simulations are ideal for this project. We welcome RL researchers, Python environment developers, AI systems engineers, and machine learning architects.

• Design and build custom tool-based RL environments for simulated knowledge work.
• Develop reliable interfaces that allow agents to interact with various digital tools.
• Test and refine state spaces, action spaces, and reward structures for optimal agent training.
• Commit approximately 20 hours per week to active development, testing, and code review.

• Professional experience as a reinforcement learning engineer or AI systems developer.
• Hands-on background building custom RL environments, simulations, or tool gyms.
• Strong proficiency in Python and standard reinforcement learning frameworks.
• Availability to contribute approximately 20 hours per week to the project.

• You will be engaged as an independent contractor.
• This is a fully remote opportunity that can be completed on your own schedule.
• Opportunities can be extended, shortened, or concluded early depending on needs and performance.
• Your participation will not involve access to confidential or proprietary information from any employer, client, or institution.
• Payments are processed weekly based on services rendered.
• We are unable to support H1-B or STEM OPT candidates at this time.