Description

You will design and implement reinforcement learning agents to solve complex sequential decision-making problems.

This role is remote.

Responsibilities

• Implement agents that maximize cumulative reward through sequential decision-making.
• Design and manage simulated environments using OpenAI Gym or Unity ML-Agents.
• Tune models to balance exploration versus exploitation trade-offs.
• Apply stability techniques, reward normalization, and experience replay during long training runs.

Required Skills

• 5+ years of relevant experience in data science or machine learning.
• Deep understanding of Markov Decision Processes (MDPs), policy/value functions, and Bellman equations.
• Proficiency with RL algorithms: Q-learning, SARSA, DQN, REINFORCE, PPO, A3C, DDPG, SAC.
• Experience with RL-specific libraries: Stable Baselines3, Ray RLlib, TensorFlow Agents, OpenAI Baselines.
• Ability to model environment dynamics and implement custom reward shaping.
• Strong foundation in Python and numerical computing.

Preferred Skills

• Experience with large-scale distributed training.
• Publications or open-source contributions in reinforcement learning.

Key Skills

Education

Any Gradute