Senior Reinforcement Learning Decisioning Engineer

Relha LLC; Boston, Northern (MA, KY); Hybrid; USD 129,000 - 178,000; Full time

Job summary

Humana is seeking a skilled Decision Intelligence Engineer to design, train, and improve the reinforcement learning policy at the heart of its Next Best Action platform. You will own the full RL lifecycle from feature engineering through production deployment for 8 million members.

The role blends research and hands-on engineering: implementing RL algorithms, building training pipelines in Databricks, and collaborating with data and platform teams to ensure safety, explainability, and

Qualifications

• Bachelor's degree in computer science or related field.
• 8+ years of software engineering experience on data-intensive platforms.
• 3+ years implementing reinforcement learning in production.
• Proficiency in Python and ML/DL frameworks (PyTorch or TensorFlow).
• Experience with distributed training and ML tooling (MLflow, Databricks).

Responsibilities

• Design, implement, and evaluate RL algorithms for healthcare decisioning (PPO, A3C, DQN, Q-learning, offline RL).
• Define member state and action spaces; onboard new programs and data sources.
• Encode clinical eligibility and constraints into the RL objective via reward shaping.
• Manage exploration-exploitation tradeoffs in production.
• Build simulation and backtesting environments to evaluate policy quality.
• Diagnose RL failures (policy collapse, credit assignment, distributional shift).
• Instrument training with MLflow; track hyperparameters and rewards.
• Own Databricks training workflow: feature engineering, normalization, distributed RL with Ray RLlib.
• Collaborate with Data Engineering to ensure correct inputs and reproducible pipelines.
• Produce production-quality PySpark feature engineering jobs; maintain data lineage.

Skills

Python

PyTorch

TensorFlow

Ray RLlib

Databricks

PySpark

MLflow

Delta Lake

Reinforcement learning

Education

Bachelor's degree in computer science or related field

Tools

PettingZoo

Gymnasium

Kafka

OpenTelemetry

Unity Catalog

Job description

Humana is seeking a skilled Decision Intelligence Engineer to design, train, and improve the reinforcement learning policy at the heart of its Next Best Action platform. You will own the full RL lifecycle from feature engineering through production deployment for 8 million members.

The role blends research and hands-on engineering: implementing RL algorithms, building training pipelines in Databricks, and collaborating with data and platform teams to ensure safety, explainability, and