Engineering Manager, Serverless Compute Platform

Impact

• Own a 0→1 service with platform-wide blast radius. Architect and launch the Execution Sandbox Service from inception to production scale. This greenfield provisioning layer will power all non-Spark compute workloads on Serverless (Notebooks, AI Agents, Remote UDFs).
• Unify a fragmented compute surface. Converge disparate CPU and GPU cluster management paths into a single provisioning service, eliminating parity bugs and enabling consistent product experiences.
• Collaborate across 5+ partner organizations. Drive alignment on API contracts and shared milestones across Serverless Platform, AI Runtime, Lakeguard, and product teams.
• Shape product strategy through deep technical understanding. Partner with Product Management to leverage this new sandbox primitive for future offerings like serverless command execution APIs and FaaS-style workloads.

Responsibilities

• Own the end-to-end delivery of the Execution Sandbox service and the engineers building it.
• Build out the full vision, guide evolution, and scale the team.
• Ensure strong execution health and that the service launches with production-grade reliability spanning a range of use cases, e.g. GPU onboarding, UDF generalization, and managed REPL.
• Manage and elevate a team of strong L3-L5 engineers, establishing clear ownership boundaries and architectural doctrine.
• Hire 2-3 additional engineers to support this expanded scope.

Requirements

• 5+ years managing engineers building and operating distributed systems in production, ideally control-plane or orchestration services.
• BS or higher in Computer Science or a related field. Equivalent practical experience is equally valued.
• Deep technical fluency in infrastructure systems. Ability to deeply review architecture docs, challenge design tradeoffs (e.g., state machine design, API boundaries), and coach senior ICs.
• Experience with multi-cloud or multi-region service deployment (AWS, Azure, GCP).
• Bias toward operational rigor. Deep commitment to observability, SLOs, pre-mortems, and healthy on-call cultures.
• Build and scale a high-caliber team.