Lead Decision Intelligence Engineer - NBA
Core
Design, train, and continuously improve reinforcement learning policies for Humana's Next Best Action platform to optimize healthcare decisions for 8 million members.
Role type
Senior IC reinforcement learning engineer (healthcare decisioning)
Builds
Production RL policies integrated into a real-time decisioning platform
Domain
Healthcare + Reinforcement Learning
Deliverable
production ML models
Required skills
Reinforcement learning (PPO, A3C, DQN, Q-learning, CQL, Decision Transformer), Python 3.x, PyTorch or TensorFlow, Ray RLlib, Databricks, PySpark, Delta Lake, MLflow, distributed training, reward shaping, constraint mapping, simulation/backtesting environments
Preferred skills
Multi-agent RL (PettingZoo), probabilistic modeling, Markov Decision Processes, linear programming, Gymnasium, Kafka feedback loops, OpenTelemetry
Technologies
Ray RLlib, Databricks, PySpark, Delta Lake, MLflow, PettingZoo, Gymnasium, Kafka, Redis, Drools, Kafka, Databricks Delta Live Tables
Responsibilities
Design and implement RL algorithms for long-horizon healthcare decisioning; define member state representation and action spaces; build simulation and backtesting environments; own nightly Databricks training workflows; manage model artifacts and lifecycle in MLflow; enforce hard business rules as constraints within RL objectives; document model behavior for clinical stakeholders.
Seniority
Senior, hands-on IC
