Principal Supercomputing Operations Software Engineer

Microsoft Azure’s Artificial Intelligence and High Performance Computing (AI/HPC) organization powers some of the world’s largest cloud native supercomputers used for frontier AI training, scientific computing, and large scale distributed simulations. Our team builds and operates hyperscale GPU clusters that consistently place Azure among global leaders in the Top500, MLPerf, and Graph500 benchmarks. By joining us, you step into the engineering core responsible for ensuring these systems remain reliable, performant, and ready for the next wave of AI innovation.

At this scale, interconnect fabrics are a first order reliability system that directly determines GPU availability, training throughput, and customer SLAs. As a Principal Supercomputing Operations Engineer, you serve as the technical authority and strategic owner for interconnect fabric operations across flagship AI supercomputing environments. You treat InfiniBand and GPU interconnect fabrics as a single end to end reliability domain, defining how they are operated, debugged, hardened, and scaled in production. This is a hands on, production first leadership role operating at the intersection of architecture, live operations, and reliability engineering.

You will lead the most complex and impactful fabric related incidents, making high stakes technical decisions under ambiguity while balancing availability, risk, long term correctness, and customer impact. Beyond resolving incidents, you define failure models, operational strategy, and systemic prevention mechanisms that reduce recurrence at fleet scale. Your impact multiplies through technical leadership: setting operational standards, influencing engineering direction across teams, mentoring senior engineers, and partnering deeply with platform, hardware, firmware, and service teams to drive durable reliability improvements.

You will architect and drive automation, diagnostics, and telemetry that materially improve operability and debuggability of interconnect fabrics, and author authoritative playbooks, TSGs, and escalation models relied on across the organization. Through your judgment, designs, and operational strategy, Azure’s largest AI platforms scale safely, predictably, and sustainably to meet the demands of next generation AI workloads.

Serve as the technical authority and DRI for InfiniBand and GPU interconnect fabric operations across large scale AI supercomputing environments, ensuring sustained GPU availability, training stability, and SLA compliance

Lead and orchestrate complex, high severity fabric incidents end to end, including detection, triage, mitigation, recovery, and root cause analysis, making high impact decisions under ambiguity

Perform deep, multi layer systems debugging across InfiniBand, Subnet Manager, GPU interconnect, PCIe, GPUs, firmware, drivers, and OS layers to identify true root causes at fleet scale

Drive operational excellence and systemic prevention by identifying recurring failure patterns, defining reliability models and failure domains, and authoring authoritative TSGs, playbooks, and escalation frameworks adopted across teams

Architect and drive automation, telemetry, diagnostics, and tooling that materially improve detection, observability, debuggability, and mean time to mitigation, raising the operational bar for interconnect fabrics across the platform