Principal ML Platform Engineer

You will design and improve systems for model training, evaluation, serving, and deployment. You will build reliable, scalable, cost-efficient infrastructure and internal tools; improve GPU and cloud workload scheduling, monitoring, and debugging; reduce operational overhead through automation and agentic systems; and make architectural trade-offs with researchers and product engineers.

Responsibilities

• Design and improve systems for model training, evaluation, and production serving
• Build reliable, scalable, and cost-efficient ML infrastructure and tooling
• Develop internal tools and workflows for humans and agents
• Design architecture for model deployment, serving, and operations
• Improve GPU and cloud workload scheduling, monitoring, and debugging
• Develop agentic systems that reduce operational overhead
• Improve observability, automation, reliability, and developer experience
• Collaborate with researchers and product engineers on platform capabilities
• Contribute technical direction and architectural trade-offs

Requirements

• Experience building or operating reliable, scalable, and maintainable production systems
• Systems thinking for bottlenecks, failure modes, interfaces, resource usage, and operability
• Experience with cloud infrastructure, Linux, and infrastructure automation
• Experience with Kubernetes and distributed production workloads
• Strong coding skills in Python or similar backend and tooling languages
• Experience building internal platforms, developer tooling, or infrastructure abstractions
• Ability to take ownership of ambiguous technical problems