Staff ML Platform Engineer
About this role
The ML Platform team at Datavant is building the infrastructure that takes models from notebooks to production without requiring teams to rebuild them each time, including training and inference across SageMaker and Databricks, self-service LLM endpoint serving, and pipelines for clinical AI products. Operating within the Data & ML Platform organization, the incoming engineer will own architectural problems, guide technical direction for the paved road, and move the group from bespoke plumbing to a cohesive, self-serve platform. Ideal candidates possess extensive engineering backgrounds with enterprise-scale ML systems, demonstrate strong technical judgment under ambiguity, and naturally incorporate AI-native coding assistants into daily workflows while handling sensitive health data responsibly.
What you will do
- Set technical direction across machine learning training, serving, and observability while acting as the final escalation point for infrastructure challenges.
- Own and evolve the paved road framework, including the shared CI/CD spine, model workflow scaffolding, and Databricks Asset Bundles.
- Lead architecture for large language model endpoint serving across managed providers and self-hosted deployments.
- Establish standards and tooling for MLflow, model registries, training image supply chains, and system observability.
- Partner with platform teammates to present a unified platform to data science, application development, and operations teams.
- Mentor senior engineers, guide platform consumers, and write high-leverage code and infrastructure-as-code alongside team members.
What you need
- Ten or more years of software engineering experience, including at least three years designing and operating enterprise-scale machine learning platforms in production.
- Hands-on production experience with Databricks or Amazon SageMaker, MLflow or equivalent tracking systems, and core machine learning frameworks like PyTorch or TensorFlow.
- Fluency in Java or JVM equivalents and Python, backed by real depth in Apache Spark for large-scale data and distributed compute.
- Extensive background in AWS networking, identity management, GPU compute, storage, and messaging services.
- Proficiency with Terraform, containerization, Kubernetes, and GitHub-based CI/CD pipelines for machine learning workloads.
- Direct experience serving large language models in production while managing costs, evaluation harnesses, and sensitive prompt routing.
- AI-native working style utilizing tools like Claude Code, Cursor, or Copilot daily with an understanding of safe application practices.
Nice to have
- Prior technical leadership on a healthcare or regulated-industry machine learning platform under HIPAA, HITRUST, or SOC 2
- Direct experience with Databricks Asset Bundles, Unity Catalog, and open table formats such as Iceberg or Delta
- Production experience with specialized inference pipelines like document understanding, computer vision, or streaming inference
- GPU capacity planning experience across strategic, tactical, and operational horizons
- Background contributing to open-source machine learning infrastructure or publishing on production systems
Skills and tech stack
Questions candidates ask
What is the remote work policy for this position?
The position is listed as remote within the United States, with coordination occurring in remote and asynchronous settings.
What is the salary range for this role?
The estimated total cash compensation range for this role is $224,000 to $280,000 USD, with individual rates differing based on level, responsibilities, skills, and experience.
What kind of tech stack and AI tools are expected?
The team works with AWS, Databricks, SageMaker, Spark, Terraform, Kubernetes, and MLflow, and expects a daily workflow utilizing AI coding assistants like Claude Code, Cursor, or Copilot.
Does this role qualify for employment visa sponsorship?
No, this job is not eligible for employment sponsorship.