Job Description



Must Have Technical/Functional Skills



Experience:

3+ years building and deploying enterprise scale decision systems. Hands-on experience with implementing policy gradient methods (PPO, A3C), value-based approaches (DQN, Q-learning) and off-policy algorithms. Deep familiarity with the Bellman equation, reward shaping, exploration-exploitation tradeoff, constraint mapping and knowing common failure points of real-world reinforcement learning systems. Ability to diagnose issues with policy learning and collapse, credit assignment issues, and distributional shifts affecting performance of the model.

Key Skills:

Deep learning frameworks (tensorflow, pytorch), linear programming, Markov decision processes, Actor-Critic methods, Offline RL methods (CQL, Decision Transformer), probabilistic modeling, databricks, Ray RLLib, Gymnasium, PettingZoo (MARL).