Senior Decision Intelligence Engineer

Humana

1d•$106,900 - $147,000•Remote

About The Position

The Senior Machine Learning Engineer, Decision Intelligence is a hands-on individual contributor responsible for building, deploying, and operating machine learning and decisioning pipelines for the NBA Decision Intelligence Platform. The role focuses on production pipeline development, MLOps, feature engineering, scoring workflows, monitoring, and optimization-aware decisioning. The engineer helps ensure the platform selects appropriate actions for members while respecting clinical eligibility, suppression rules, channel constraints, program goals, and operational capacity. The role works closely with machine learning, data, and platform engineers, product owners, and decision engine teams to deliver reliable, scalable, and auditable production systems. The position is remote or hybrid, preferably based in Boston, Massachusetts. Typical working hours are Monday through Friday, eight hours per day, with some flexibility depending on business needs. Occasional travel to Humana offices for training or meetings may be required. Home or hybrid employees must have internet service with at least 25 Mbps download and 10 Mbps upload and a dedicated workspace that protects member PHI/HIPAA information.

Requirements

Bachelor's degree in computer science or a relevant field.
At least 5 years of post-undergraduate software engineering or quantitative research experience building and operating large-scale production systems, emphasizing data-intensive platforms, recommendation systems, optimization engines, or simulation frameworks serving millions of users.
At least 2 years of post-graduate software engineering or quantitative research experience building and operating large-scale production systems, emphasizing data-intensive platforms, recommendation systems, optimization engines, or simulation frameworks serving millions of users.
At least 2 years of hands-on experience implementing reinforcement learning, operations research methods, or simulation-driven decision systems in production. Relevant backgrounds include policy-gradient and value-based reinforcement learning (PPO, A3C, DQN, CQL), stochastic dynamic programming, discrete-event simulation, or large-scale combinatorial or constrained optimization.
Deep familiarity with Markov Decision Processes, Bellman-equation-based value estimation, reward or objective shaping, exploration-exploitation tradeoffs, and constraint formulation in real-world decision systems.
Demonstrated ability to diagnose failure modes in learned or optimized policies, including instability, poor credit assignment across long horizons, and distributional shift across large populations.
Proficiency in Python 3.x and experience with PyTorch or TensorFlow for policy network or learned model implementation.
Experience with Ray RLlib or equivalent distributed computation frameworks for large-scale training or optimization.
Experience with Databricks, PySpark, and Delta Lake for large-scale ML or data pipelines processing tens of millions of records.
Experience with MLflow for experiment tracking, model registry, and artifact management.
Experience shipping systems that operate reliably under production load, rather than only research or prototype work.

Nice To Haves

Experience with multi-agent reinforcement learning frameworks such as PettingZoo, or equivalent multi-agent simulation and coordination methods.
Familiarity with operations research methods for constrained sequential decisioning, such as linear programming, mixed-integer programming, Lagrangian relaxation, or constraint programming.
Experience operating decision or optimization systems in regulated domains such as healthcare, finance, or insurance, where member safety, auditability, and explainability are requirements.
Experience building simulation environments using Gymnasium, SimPy, AnyLogic, or equivalent frameworks for policy evaluation and backtesting.
Familiarity with event-driven feedback loops and how disposition signals feed retraining or re-optimization pipelines.
OpenTelemetry instrumentation experience for ML or optimization pipeline observability.

Responsibilities

Build, deploy, and operate machine learning and decisioning pipelines for the NBA Decision Intelligence Platform.
Develop production pipelines, MLOps capabilities, feature engineering, scoring workflows, monitoring, and optimization-aware decisioning.
Help ensure decisions respect clinical eligibility, suppression rules, channel constraints, program goals, and operational capacity.
Deliver reliable, scalable, and auditable production systems in collaboration with ML engineers, data engineers, platform engineers, product owners, and decision engine teams.