ML Infrastructure Engineer

Responsibilities

• Own the architecture, implementation, reliability, and evolution of Maven's machine learning infrastructure.
• Build backend services and platforms for managing data, artifacts, jobs, logs, metadata, and compute resources across cloud and on-premise environments.
• Design scalable systems for workload orchestration, storage, observability, security, and infrastructure automation.
• Build intuitive internal tools and abstractions that make complex infrastructure easy for engineers to use.
• Lead technical and commercial discussions with cloud and ML compute providers, including capacity planning, performance, reliability, and cost.

Requirements

• Significant experience designing, building, and operating production backend, distributed, or compute infrastructure.
• A track record of independently owning complex infrastructure projects from architecture through deployment and ongoing operation.
• Strong programming ability in Python, Go, Rust, C++, or a similar backend or systems language.
• Experience operating GPU compute infrastructure and orchestrating distributed workloads using Kubernetes, Ray, ZenML, or similar systems.
• Experience designing and operating storage systems, observability platforms, infrastructure-as-code, and secure access controls.
• Experience managing large-scale GPU fleets or hybrid cloud and on-premise compute environments.
• Experience building internal developer platforms, CLIs, SDKs, or other self-service infrastructure tools.
• Strong technical judgment, leadership, and communication skills, with the ability to drive decisions across teams and external partners.
• Self-starter attitude with the ability to identify priorities and deliver durable solutions in a fast-paced startup environment.

Nice-to-Haves

• Familiarity with GPU architecture, accelerator-aware software design, and profiling compute-intensive workloads.
• Exposure to infrastructure supporting large-scale robot learning workloads, including policy training, simulation, and multimodal data pipelines.
• Familiarity with SOC 2 controls, security practices, and audit readiness.