Job Summary

AuraCare Technologies is seeking an AI Healthcare Data Scientist to build and deploy machine learning models that analyze continuous remote patient monitoring (RPM) data. In this role, you will transform raw physiological streams (such as heart rate, oxygen saturation, and blood pressure) into actionable, predictive clinical insights. Your work will directly empower clinicians to intervene before health crises occur, lowering hospital readmission rates and saving lives.

Key Responsibilities

• Model Development: Design, train, and validate predictive machine learning and deep learning models optimized for time-series physiological data.
• Data Pipeline Architecture: Partner with data engineers to build robust, scalable pipelines capable of handling massive streams of real-time health metrics.
• Clinical Validation: Collaborate with the Chief Medical Officer to ensure models align with clinical guidelines and minimize false-alarm fatigue for medical staff.
• Bias & Fairness Mitigation: Implement rigorous testing frameworks to identify and eliminate algorithmic bias across diverse demographic patient populations.
• Deployment & Monitoring: Deploy models into production environments and continuously monitor performance, drift, and accuracy over time.

Minimum Qualifications

• Education: Master’s or Ph.D. in Data Science, Computer Science, Biomedical Engineering, or a highly quantitative field.
• Experience: 3+ years of professional experience building production-level machine learning models, specifically within the healthcare, biotech, or life sciences sectors.
• Technical Skills: Mastery of Python and standard data science libraries (TensorFlow, PyTorch, Scikit-Learn, Pandas, SQL).
• Domain Knowledge: Hands-on experience working with protected health information (PHI), clinical datasets, and EHR integration concepts.

Preferred Skills

• Familiarity with HIPAA compliance, HITRUST frameworks, and FDA regulations for Software as a Medical Device (SaMD).
• Experience processing real-time streaming data via Kafka or AWS Kinesis.