Lead Decision Intelligence Engineer

This is a fully remote position, open to applicants in United States.

📋 Description

• Design, implement, and assess reinforcement learning algorithms for healthcare decision-making involving long horizons and sparse rewards.

• Establish and uphold member state representations and action spaces.

• Utilize the Bellman equation, reward shaping, and constraint mapping to encode clinical eligibility criteria and program goals.

• Oversee exploration-exploitation tradeoffs within a production healthcare context.

• Create simulation and backtesting environments leveraging historical member journey data.

• Identify and resolve reinforcement learning failure modes, including policy collapse, credit assignment errors, and distributional shifts.

• Set reward thresholds and automated evaluation gates within nightly Databricks workflows.

• Block the promotion of underperforming policies to the MLflow production environment.

• Track training runs using MLflow instrumentation.

• Manage nightly Databricks training workflows, feature engineering, distributed reinforcement learning training, and batch scoring for 8 million eligible members.

• Develop production-quality PySpark feature engineering jobs and ensure data lineage via Databricks Unity Catalog.

• Oversee model artifacts, versioning, lifecycle management, and rollback functionalities within the MLflow Model Registry.

• Implement multi-agent reinforcement learning when coordination within households or populations is necessary.

• Enforce constraints related to member caps, cooldown periods, and clinical eligibility.

• Collaborate with stakeholders from the Rules Engine, Data Engineering, Decision Engine, platform architecture, clinical, and compliance teams.

• Integrate model outputs with real-time decision-making processes and Redis-cached recommendations.

• Define feedback loop contracts based on disposition outcomes through Kafka and Databricks Delta Live Tables for retraining purposes.

• Document model behaviors, limitations, and potential failure modes.

• Support explainability requirements for decisions affecting members.

• Leverage AI-assisted engineering tools for scaffolding, testing, and documentation while ensuring that core model logic remains human-authored and peer-reviewed.

⛳️ Requirements

• Bachelor's degree in computer science or a related discipline.

• Over 8 years of software engineering experience in constructing and managing large-scale production systems.

• Focus on data-intensive platforms, recommendation systems, or optimization engines serving millions of users.

• At least 3 years of direct experience in implementing reinforcement learning or deep learning systems in a production environment.

• Familiarity with policy gradient methods such as PPO and A3C, value-based techniques including DQN and Q-learning, or offline RL algorithms like CQL and Decision Transformer.

• Strong understanding of the Bellman equation, reward shaping, exploration-exploitation tradeoffs, and constraint mapping.

• Capability to diagnose issues related to policy collapse, credit assignment, and distributional shifts.

• Proficient in Python 3.x.

• Experience with PyTorch or TensorFlow.

• Familiarity with Ray RLlib.

• Experience using Databricks, PySpark, and Delta Lake.

• Knowledge of MLflow.

• Proven track record of delivering reliable ML systems under production loads.

• This role is not available for work visa sponsorship.

• Minimum home internet speed of 25 Mbps for downloads and 10 Mbps for uploads.

• A dedicated workspace free from ongoing interruptions to safeguard PHI/HIPAA information.

🏝️ Benefits

• Bonus incentive plan based on both company and/or individual performance.

• Medical, dental, and vision coverage.

• 401(k) retirement savings plan.

• Paid time off.

• Company holidays and personal days.

• Paid parental and caregiver leave.

• Short-term and long-term disability insurance.

• Life insurance coverage.

• Flexible working hours may be available based on business requirements.

• Option for remote work arrangements.

• Occasional travel to Humana offices for training sessions or meetings.

• Requirement for a dedicated home workspace to protect member PHI/HIPAA information.

• Home internet service requirements with potential upgrade support if needed.