SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Metriport is building the infrastructure layer for US healthcare by solving medical data exchange—one of the biggest unsolved problems in the healthcare system. The company connects to thousands of legacy data sources, transforms raw unusable data into clean, standardized formats, and delivers actionable intelligence to care teams. With $28.4M in funding from top-tier VCs (Matrix, ARTIS, Y Combinator), 100+ customers including Amazon One Medical and Circle Medical, and multi-million dollar ARR, Metriport has achieved product-market fit and is scaling rapidly.
This is the first Machine Learning Engineer role at Metriport. You will own machine learning end-to-end: from problem framing with customers, to training models, to serving them in production and monitoring their performance. This is applied ML on messy, high-dimensional, real-world healthcare data—not research, not LLM-wrapping.
Day-to-day responsibilities include:
- Building predictive models on clinical data at scale (risk stratification, utilization prediction, care-gap detection)
- Owning the full ML lifecycle: problem framing, feature engineering over sparse and high-dimensional clinical data, training, evaluation, deployment, and monitoring for drift and degradation
- Turning unstructured clinical data (PDFs, images, doctor notes) into structured, usable records
- Standing up ML infrastructure: training and inference pipelines, experiment tracking, model versioning, and evaluation frameworks
- Working directly with founders, engineers, and customers to identify highest-leverage ML problems
Example projects: building models that predict expected hospitalizations from medical history, converting freeform doctor notes and scanned documents into structured data, or building classifiers to categorize billions of clinical documents with incomplete metadata.
The company operates with high autonomy and minimal bureaucracy. Leadership is in the office six days a week and generally expects the team to be available six days a week—not for hours-counting, but because there's always more to build when revolutionizing healthcare. The team is small, high-output, mostly former founders including YC alumni, and hires based on competence. Time off is flexible and never denied.
Tech stack: Node.js and TypeScript for core business logic, Python for data and ML workflows, AWS infrastructure (ECS, Lambda, SQS, SNS, Batch, CDK), data in S3, PostgreSQL/Aurora, DynamoDB, Snowflake, and a FHIR server, with Athena for querying and SageMaker for analytics/ML.
REQUIREMENTS:
- 7+ years building ML systems that run against real data at scale
- Strong grounding across the ML spectrum: regression and tree-based methods, feature engineering and dimensionality reduction, deep learning. Must have built and trained models yourself, not just orchestrated APIs
- Track record of ML delivering measurable results in production
- Strong software engineering fundamentals: built or operated large-scale backend systems on the cloud (ideally AWS), can own training pipelines, model serving, and monitoring end-to-end
- Strong data skills: SQL, working with large datasets, building pipelines that feed models
- Located in San Francisco or Bay Area (or willing to relocate)
- Bonus: Healthcare standards/technologies (FHIR, HIE, IHE, EHR/EMR, NPI, TEFCA, ADT, HL7, HEDIS, RAF, SNOMED, LOINC, ICD-10); founder experience or being first/only ML hire at early-stage startup
