Machine Learning Engineer

Role Overview

The Machine Learning Engineer owns the “intelligence layer” behind clinical agents, ensuring they provide clinically meaningful recommendations rather than simple rule execution. You will design, train, validate, and monitor ML models that power Markov state transitions, clinical decision support, patient risk stratification, adherence prediction, and dosage optimization—building systems that are explainable, safe, and auditable for regulators.

Responsibilities

• Build evaluation & validation frameworks for agent-driven clinical recommendations, including:
  • safety guardrails
  • confidence thresholds
  • human-in-the-loop escalation triggers
• Develop patient prediction models for:
  • adherence prediction
  • adverse event likelihood
  • dosage titration optimization
  • churn/dropout risk
• Implement predictive analytics pipelines on AWS, including:
  • Amazon Forecast (DeepAR+) for time-series clinical predictions
  • S3 Vectors for embedding-based patient similarity/retrieval
  • Bedrock for agent inference
• Design and build RAG architecture to ground agent responses in clinical protocols and formulary data
• Own model lifecycle management:
  • training pipelines
  • feature stores
  • model versioning
  • A/B testing
  • drift detection and retraining triggers
• Build explainability layers for clinical recommendations
• Collaborate with clinical teams to convert protocols/pharmacy domain knowledge into:
  • features
  • labels
  • validation criteria
• Establish monitoring & alerting:
  • prediction quality dashboards
  • distribution shift detection
  • automated alerts when performance drops below clinical safety thresholds
• Ensure ML decisions are logged and reproducible in coordination with DevOps/AgentOps for regulatory review

Requirements

• Applied ML / statistical modeling (expert)
  • Production experience shipping ML models (not only notebooks)
  • Deep experience with classification, regression, time-series forecasting, and probabilistic graphical models (e.g., Markov chains, HMMs, Bayesian networks)
  • Strong experimental design skills and A/B testing methodology; statistical significance in clinical contexts
• NLP & LLM engineering (expert)
  • Production experience with LLM-based systems (prompt engineering, fine-tuning, RAG design, embeddings)
  • Understanding of hallucination mitigation and grounding techniques
  • Ability to constrain outputs to clinically safe boundaries
  • Bedrock experience preferred (Claude, Titan), with equivalent depth in OpenAI/Vertex AI/Azure OpenAI acceptable
• AWS ML services (strong)
  • Hands-on experience with at least 3 of:
    • SageMaker (training, endpoints, pipelines)
    • Bedrock (agents, knowledge bases, model invocation)
    • Amazon Forecast (DeepAR+)
    • S3 Vectors
    • Comprehend Medical
    • HealthLake analytics
  • Comfortable building end-to-end ML pipelines on AWS
• Feature engineering & data pipelines (strong)
  • Experience building feature stores and training data pipelines from healthcare sources
  • Must understand FHIR resource structures to extract features from Patient, Observation, Condition, MedicationRequest, and related resources

Nice to Have

• Experience specifically with AWS Bedrock and its healthcare-oriented capabilities (preferred mentioned)