4 days ago
Remote, United StatesStaff+
Responsibilities
- Set technical direction for ML training, serving, and observability and act as the escalation point for complex infrastructure issues.
- Own the shared paved-road framework, CI/CD spine, model-workflow scaffolding, and Databricks Asset Bundles.
- Lead architecture for LLM endpoint serving across Databricks, AWS, Snowflake, and self-hosted deployments, including latency, cost, caching, evaluation, and PHI-safe routing.
- Own standards and tooling for MLflow, model registry, training image supply chains, and training and inference observability.
- Partner with Data & ML Platform, Data Science, App Dev, and Operations teams to provide a cohesive ML platform.
- Contribute technical guidance to vendor and platform selection decisions.
- Mentor senior engineers, advise platform consumers, and write high-leverage code and infrastructure-as-code.
Requirements
- 10+ years of software engineering experience, including 3+ years designing, evolving, and operating enterprise-scale ML platforms in production.
- Strong technical judgment, experience setting standards and influencing peers, and the ability to lead through ambiguity.
- Production experience with Databricks and/or Amazon SageMaker, MLflow or an equivalent tracking and registry system, and a core ML framework such as PyTorch or TensorFlow.
- Fluency in Java or a JVM equivalent and Python, with deep Apache Spark experience for large-scale data and distributed computing.
- Deep AWS experience including networking, IAM, GPU compute, storage, and messaging services.
- Fluency with Terraform, containers, Kubernetes, and GitHub-based CI/CD for ML workloads.
- Direct production experience serving LLMs, including cost management, evaluation harnesses, and safe handling of sensitive prompts and outputs.
- Daily use of Claude Code, Cursor, Copilot, or equivalent AI coding tools, with the ability to apply them responsibly to sensitive data.
- Clear written and verbal communication, particularly in asynchronous and remote settings.
- Preferred qualifications include healthcare or regulated-industry ML platform leadership, HIPAA/HITRUST/SOC 2 experience, Databricks Asset Bundles, Unity Catalog, Iceberg, Delta, specialized inference pipelines, GPU capacity planning, Kafka, Kinesis, AI evaluation and red-teaming, and open-source or published ML infrastructure work.
Benefits
- The role is not eligible for employment sponsorship.
- Post-offer health screenings and proof or completion of required vaccinations may be required depending on client and state requirements, with case-by-case exemption review.
- Datavant offers total rewards as part of its employee benefits strategy.
- The employer provides equal employment opportunity and reasonable workplace accommodations.
Tech Stack
Categories
About Datavant
Datavant builds a healthcare data collaboration platform and network that enables privacy-preserving exchange, linkage, and Release of Information across providers, payers, life sciences, and researchers. It sells software and data services for interoperability, de-identification, and compliant record retrieval, used to route more than 60 million health records among thousands of organizations. Privately held and headquartered in New York City, it reports working with 75% of the 100 largest U.S. health systems and 350+ real-world data partners.