Principal Machine Learning Engineer

Job Title:

Principal Machine Learning Engineer Role Overview

A rapidly growing healthcare AI company is transforming how clinicians monitor and care for vulnerable populations. The organization builds predictive health technology that combines contactless sensing devices with advanced machine learning to detect early signs of patient deterioration. Their platform analyzes physiological signals and clinical data to help healthcare providers intervene earlier and prevent avoidable hospitalizations.

With thousands of patients monitored daily across post-acute care environments, the company is expanding its engineering team to further advance the predictive models at the core of its platform. This role sits at the intersection of applied machine learning, healthcare data, and real-world deployment.

The Principal Machine Learning Engineer will own the end-to-end lifecycle of predictive models that power clinical decision support and operational workflows used in production environments. This individual will contribute to both improving existing risk prediction models and exploring new applications of machine learning across clinical and biometric datasets.

This is a hands‑on, high‑impact role for someone who enjoys solving complex ML problems with real‑world consequences, building models that must perform reliably on messy real‑world data, and rapidly iterating in a startup environment where shipped models directly affect patient outcomes.

• Design, train, and continuously improve production‑grade machine learning models for predictive risk scoring, clinical classification, and health deterioration detection
• Apply statistical learning approaches including gradient boosting methods (such as XGBoost, Light
  
  GBM, Cat Boost) as well as modern deep learning approaches including transformer-based architectures where appropriate
• Work with time‑series and longitudinal datasets derived from physiological signals, vital signs, and operational healthcare data
• Design experiments to evaluate new modeling techniques, feature engineering strategies, and training approaches that improve predictive performance
• Own the full model lifecycle from research and experimentation through validation, production deployment, monitoring, and iteration
• Develop and maintain feature pipelines that transform raw sensor data, clinical indicators, and behavioral signals into model‑ready datasets
• Collaborate closely with clinicians, engineers, and product stakeholders to ensure models are interpretable, clinically useful, and aligned with real‑world workflows
• Contribute to exploration of new AI capabilities, including applications of large language models (LLMs) for clinical documentation and workflow optimization
• Investigate new signal sources and data modalities that may improve prediction accuracy or enable new product capabilities
• Produce explainability outputs (such as SHAP or feature attribution) to support transparency, auditing, and trust in model predictions
• Partner with engineering teams to deploy models into production systems through APIs and scalable pipelines
• Measure real‑world impact of models using operational and clinical outcome metrics
• Contribute technical leadership in shaping modeling direction and future ML team expansion

• 5–10+ years of experience developing and deploying machine learning models in production environments
• Strong hands‑on experience applying statistical and machine learning techniques to real‑world datasets
• Experience improving model performance through experimentation, feature engineering, or training optimization
• Advanced Python expertise and experience with ML tooling such as Num Py, pandas, scikit‑learn, PyTorch, Tensor Flow, or similar frameworks
• Strong foundation in statistics, machine learning theory, and model evaluation methodologies
• Experience working with structured, tabular, or time‑series datasets
• Demonstrated ability to own ML projects end‑to‑end, from experimentation through deployment and monitoring
• Ability to communicate technical trade‑offs and model behavior to cross‑functional stakeholders
• Comfort working in ambiguous problem spaces where experimentation…