Location: REMOTE / Nashville, Tennessee
This job allows you to work remotely.
Our client is a venture-backed digital health company helping people with joint, back, and muscle (musculoskeletal) conditions get to the right specialty care. The company builds a full view of each patient, from symptoms and history to preferences and goals, pairs it with predictive data and current clinical evidence, and guides each person to the treatment path most likely to work, staying with them along the way.
The company is remote-first with an optional hub in Nashville, and is scaling quickly into new markets. The team is collaborative, curious, and driven by the mission of getting people out of pain.
This Role:
The Senior ML/AI Engineer will take models from prototype into production and keep them healthy, on a platform whose output is used daily by care teams working with real patients. The data science group runs a growing set of predictive models and a real-time decision engine. The hard part is not the prediction itself; it is choosing the right action for each patient, at the right time and through the right channel, and capturing what happened so the system keeps improving.
You will own the engineering behind the predictive and decisioning systems: deployment and retraining workflows, monitoring, the real-time event paths that carry model output to the people and systems that act on it, and the GenAI tools care teams already rely on. You will help move the platform from a Databricks lakehouse toward a hybrid setup on AWS, and raise the engineering bar for a team of strong modelers and healthcare data scientists. The role reports to the company's AI leader and works closely with data engineering, software engineering, clinical, and operations partners.
You Will:
• Build the layer between model score and action: action selection logic, holdout and experiment assignment, outcome capture, and pipelines that measure lift by action and channel
• Design and build real-time, event-driven integrations from model output to downstream systems: streaming pipelines, versioned schemas and contracts, replay, ordering, and failure handling
• Track and control cost to serve: instrument compute, streaming, inference, and LLM spend per model and per decision, set budgets and alerts, and treat cost as a release criterion next to latency and accuracy
• Deploy, monitor, and retrain production models with CI/CD, model registry, and alerting built to a healthcare reliability standard, including drift, coverage, and calibration by patient group and market
• Build and improve predictive models on healthcare claims, authorization, and operational data, and validate them under production conditions before release
• Drive the AWS build-out alongside the Databricks lakehouse: infrastructure as code, IAM and secure data handling, containerized inference, and managed foundation-model services under PHI controls
• Extend the feature store toward real-time features as new event sources come online
• Own and extend production GenAI applications, including LLM pipelines and agentic workflows, with evaluation, retrieval, guardrails, and cost controls suited to patient data
• Partner with data scientists to productionize new models, and with data and software engineering on shared pipelines and interfaces
• Help set and uphold engineering standards with the ML engineering lead (code review, automated testing, reproducibility, documentation, release discipline) and coach data scientists on production practice
• Travel occasionally (up to roughly 10%) for onsite meetings, team sessions, and company events
Must Have Skills:
• 5+ years in ML engineering, MLOps, or data science, including 3+ years running ML systems in production
• Production experience with Databricks, PySpark, Delta Lake, and MLflow
• Production experience with AWS data and ML services and infrastructure as code (Terraform, CDK, or CloudFormation)
• Hands-on design and operation of event-driven or streaming systems (Kinesis, Kafka, or similar) with schema or contract management
• Strong Python and SQL, and experience building and evaluating models with scikit-learn, XGBoost, or PyTorch
• CI/CD for ML workloads plus model and pipeline monitoring
• Bachelor's degree in computer science, engineering, or a related field, or equivalent experience
Nice to Have Skills:
• Recommendation, decisioning, or experimentation systems in production, including uplift or bandit approaches
• Healthcare data (claims, eligibility, prior authorization, EHR) and PHI handling
• dbt and lakehouse data modeling
• Containerized model serving with staged or canary deployments
• Amazon Bedrock or another managed foundation-model platform under a HIPAA BAA
• LLM applications, agentic frameworks, and retrieval systems in production, including evaluation and guardrails
