Website:
provue.ai
Job details:
We are building an applied AI lab working with frontier AI labs to train Personal Superintelligence. We build environments and evals for frontier consumer agents.
About the Role:
Provue is looking for an RL Environment Engineer to help build the environments and evaluation systems through which AI agents learn and improve.
In this role, you will work at the intersection of AI research and engineering — creating realistic environments for AI agents to interact with, designing ways to measure whether they have completed tasks correctly, and building the infrastructure needed to run these experiments reliably at scale.
You don't need to be an expert in every area of reinforcement learning. We are looking for someone who is a strong technical problem solver and is excited about building the infrastructure that enables AI agents to learn.
What You'll Do:
- Build environments where AI agents can interact with tools, applications, and simulated or real-world scenarios.
- Develop reliable, resettable environments that can be used repeatedly for training and evaluation.
- Design automated ways to determine whether an agent has successfully completed a task.
- Build task generators and evaluation systems to test agents across different scenarios.
- Develop infrastructure for running large numbers of agent rollouts and experiments.
- Create systems that make experiments reproducible, consistent, and measurable.
- Identify and investigate failure modes in agent behaviour.
- Work with researchers to improve task design, reward mechanisms, and evaluation methodologies.
- Experiment with different approaches to training and evaluating AI agents.
- Contribute to building the technical foundations for reinforcement learning and agent research.
What We're Looking For:
- 1–4 years of experience in software engineering, ML engineering, AI research, or a related technical field.
- Strong programming skills, particularly in Python.
- Experience building technical systems, automation, simulations, testing infrastructure, or developer tools.
- Strong problem-solving and debugging skills.
- Comfortable working with APIs, containers, databases, or other technical infrastructure.
- Understanding of basic machine learning or AI concepts.
- Strong interest in AI agents, reinforcement learning, and AI evaluation.
- Ability to work in an experimental environment where problems are often open-ended and require independent thinking.
Good to Have:
- Experience with reinforcement learning or RL environments.
- Experience building agent systems or applications involving LLMs.
- Familiarity with tools such as Docker, Gym/Gymnasium, PyTorch, or similar frameworks.
- Experience with automated testing, benchmarking, evaluation frameworks, or simulation environments.
- Experience designing reward functions, verifiers, or programmatic evaluation systems.
- Research experience or contributions to technical/open-source projects.
Click on Apply to know more.
