About the Company We are a digital health startup developing non-invasive, real-time hemodynamic monitoring technology for cardiovascular care. Our platform combines wearable biosensors with advanced machine learning to deliver clinically actionable insights for heart failure and vascular risk management. We're now preparing to scale our algorithms from validated prototypes to production-grade, regulatory-compliant systems. About the Role We’re hiring a Senior Machine Learning Engineer to lead the transition of our cardiovascular signal analysis models from research to real-world deployment. You’ll work closely with our CTO and clinical team to optimize model accuracy, enhance data pipelines, expand validation datasets, and prepare for regulatory submissions. You’ll also play a key role in building our MLOps infrastructure. Key Responsibilities Build and deploy ML models (classification/regression) using classical methods (DT, RF, SVMs) and deep learning (CNNs, RNNs, LSTM) Develop preprocessing pipelines for ECG, PPG, BP, and imaging data Implement signal quality metrics, ensemble optimization, and multi-site model validation Integrate multimodal clinical data: medical notes, digital twin outputs, ultrasound measurements Expand model predictions to new heart failure markers (e.g. pulmonary artery pressure, PVR) Build interpretable models aligned with clinical explainability standards Establish robust ML workflows: versioned data, automated training, reproducible validation Translate technical work into documentation for regulatory approval and clinical studies First 90 Days Deliver a working ML model on a defined patient dataset Build and document a high-quality data pipeline (ETL, labeling, validation) Establish baseline model metrics (accuracy, recall, F1, AUC) Draft clinical-grade documentation of data lineage, model assumptions, and architecture Begin compliance review aligned with HIPAA and Good ML Practices Tools & Stack Languages: Python (scikit-learn, PyTorch or TensorFlow, pandas, NumPy) Signal Processing: Custom and open-source cardiovascular signal tools Data: Custom clinical datasets, public sources (MIMIC, Mayo), multimodal records MLOps: Model versioning, reproducibility, documentation Compliance: HIPAA, SaMD, GMLP standards Requirements PhD in Biomedical Engineering, CS, or related; OR MS + 4+ years healthcare ML experience Hands-on experience developing ML models for physiological signals (ideally cardiovascular) Experience deploying models from research to production Strong background in signal processing and multimodal feature engineering Knowledge of FDA medical device regulations and SaMD compliance Fluent English and ability to work overlapping US hours Preferred Qualifications Experience with edge deployment for wearables Clinical text processing (EHR, medical notes) Background in cardiovascular physiology and biomarkers Experience with ultrasound data and vascular measurements Familiarity with FHIR/HL7, HIPAA-compliant pipelines Additional Details Work arrangement: Remote, offshore with US time zone overlap Interview process: Technical interview + management interview Soft skills: Documentation, scientific communication, cross-functional collaboration Compensation: TBD