Job Description
M365 Copilot inference is a high-impact engineering team advancing applied AI and large-scale machine learning across Microsoft. We design and operate the platform powering Microsoft 365 Copilot experiences. Our team is operating at massive GPU scale across multiple regions and SKUs in global datacenters. We build the core LLM API, routing, capacity, and control plane services that turn that fleet into Copilot experiences. We are hiring a Principal Group Software Engineering Manager to own GPU fleet health, capacity intake and planning, and automated model deployment for Copilot. This is one of the most strategic leadership roles in Copilot: every feature, experiment, and model launch flows through the systems this leader owns. You will lead existing teams, grow the org, and build the control plane that turns capacity management from a manual, ticket-driven process into an automated, self-driven platform. You will own end‑to‑end GPU fleet health and capacity platform, establishing a single source of truth with strong observability across hardware, hosts, and workloads to drive utilization and reliability. Design and scale capacity intake, planning, and deployment reducing models time‑to‑production and meeting SLAs for priority workloads through automation and data‑driven operations. Build a unified control plane that connects intake, planning, deployment, and fleet operations, enabling global optimization across cost, latency, compliance, and flexible model scaling (0→1 platform ownership). Microsoft’s mission is to empower every person and every organization on the planet to achieve more. As employees we come together with a growth mindset, innovate to empower others, and collaborate to realize our shared goals. Each day we build on our values of respect, integrity, and accountability to create a culture of inclusion where everyone can thrive at work and beyond.
