Staff ML Platform Engineer

. Set technical direction across ML training, serving, and observability .

Posted 9/25/2026full-timeRemote • United StatesLead💰 $224,000 - $280,000 per yearWebsite

Tools & technologies

ApacheAWSJavaKafkaKubernetesPythonPyTorchSparkTensorflowTerraformUnity

Key responsibilities & impact

Set technical direction across ML training, serving, and observability
Serve as final escalation point for complex infrastructure problems involving GPU capacity, Spark tuning, and production incidents
Own and evolve the paved-road framework, including shared CI/CD, model-workflow scaffolding, and Databricks Asset Bundles
Lead architecture for LLM endpoint serving across Databricks, AWS, Snowflake, and self-hosted deployments
Address latency, cost, caching, evaluation, and PHI-safe routing for LLM serving
Own standards and tooling for MLflow, model registry, training image supply chain, and observability
Partner with Data & ML Platform, Data Science, App Dev, and Operations teams
Provide technical input to vendor and platform selection decisions
Mentor senior engineers and guide platform consumers
Write high-leverage code and Infrastructure-as-Code hands-on

What you’ll need

10+ years of software engineering experience, including 3+ years designing, evolving, and operating enterprise-scale ML platforms in production
Strong technical judgment under ambiguity and track record of setting standards and influencing peers
Production experience with Databricks and/or Amazon SageMaker
Experience with MLflow or equivalent tracking and registry system
Experience with at least one core ML framework, such as PyTorch or TensorFlow
Fluency in Java or a JVM equivalent and Python
Deep Apache Spark experience for large-scale data and distributed compute
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 experience serving LLMs in production, 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
Clear written and verbal communication, especially in asynchronous remote settings
This job is not eligible for employment sponsorship
Preferred: technical leadership on healthcare or regulated-industry ML platforms; Databricks Asset Bundles, Unity Catalog, Iceberg, Delta; specialized inference pipelines; GPU capacity planning; Kafka or Kinesis; clinical or safety-sensitive AI evaluation and red teaming; open-source ML infrastructure or production ML publications

Comp & perks