Software Engineer

Orphic Health is seeking a Data Engineer to build and scale the clinical data behind its Clinical Reasoning System. The role focuses on growing, structuring, and maintaining trustworthy clinical data, supporting model training, and operating secure, reliable data infrastructure.

Responsibilities:

• Grow and maintain our clinical corpus: design and build ETL/ELT pipelines that ingest, normalize, and quality-check clinical guidelines, journal literature, and more
• Model and version clinical data: schemas, lineage, and provenance, so every answer CRS gives can be traced back to its source
• Make data quality measurable: validation, coverage and freshness monitoring, and alerting when a source drifts or breaks
• Build the datasets and pipelines behind model training: curating, labeling, and versioning training data, and running fine-tuning and evaluation jobs with our AI engineers
• Help secure our data for PHI and licensed content: access controls, encryption, retention, and audit evidence for our SOC 2 / HIPAA program
• Deploy and operate data infrastructure across SaaS and dedicated customer clouds (AWS, Azure) with infrastructure-as-code, with attention to scale, cost, and reliability
• Apply AI to how we work: coding agents, automated code review, and AI-assisted data validation

Requirements:

• You have a degree in Computer Science, Engineering, or a related field, or equivalent experience
• You have 3+ years building data-centric production systems
• You write strong Python and SQL, and you've built pipelines that process large document or data corpora reliably and incrementally
• You've built data pipelines that feed LLM, RAG, or search systems in production: parsing, chunking, embeddings, and indexing
• You're hands-on with AWS or Azure, Kubernetes, and infrastructure-as-code
• You've worked in a startup or on 0 → 1 projects, and you take ownership when things are ambiguous
• You're comfortable working directly with clinicians and customer data teams to understand what the data means
• You're AI-native: you use coding agents and AI tools daily to move faster and more reliably
• You must be authorized to work in the US without current or future visa sponsorship
• Experience with data pipeline frameworks and storage, orchestration, and vector stores, such as dlt, Delta Lake, Flyte, LanceDB
• Hands-on model training or fine-tuning, including distributed or GPU workloads
• Security & compliance: SOC 2 / HIPAA controls, cloud security hardening
• Experience in healthcare technology or another regulated industry
• Familiarity with clinical terminology and healthcare data standards such as FHIR, HL7, SNOMED CT, RxNorm, and ICD-10
• Python services and APIs that serve data to applications