Cloud Principal Hardware Engineer
- Location
- Aliso Viejo / Mountain View / Redmond
- Work model
- On-Site
- Level
- Principal
- H-1B history
- 2,066 approvals (FY2023)
- Posted
Skills
About this role
Overview
Microsoft Silicon, Cloud Hardware, and Infrastructure Engineering (SCHIE) is the team behind Microsoft’s expanding Cloud Infrastructure and responsible for powering Microsoft’s “Intelligent Cloud” mission. SCHIE delivers the core infrastructure and foundational technologies for Microsoft’s over 200 online businesses including Bing, MSN, Office 365, Xbox Live, Teams, OneDrive, and the Microsoft Azure platform globally with our server and data center infrastructure, security and compliance, operations, globalization, and manageability solutions. Our focus is on smart growth, high efficiency, and delivering a trusted experience to customers and partners worldwide and we are looking for passionate engineers to help achieve that mission. As Microsoft’s cloud business continues to grow, the ability to deploy new offerings and HW infrastructure on time, in high volume, with high quality and lowest cost is of paramount importance. To achieve this goal, the SCHIE team is instrumental in defining and delivering measures of success for hardware design, qualification, fleet support, scale, and sustainability related to Microsoft cloud hardware. Azure Memory and Storage Center of Excellence (AMS CoE) is part of the SCHIE organization focusing on memory and storage devices going into the cloud hardware servers. AMS provides memory and storage solutions to Azure and drives memory and storage suppliers to deliver high quality products that meet our requirements. AMS CoE also leads Azure Fault Prediction and Detection (AFPD), the platform that predicts, detects and repairs hardware failures across the Azure fleet. We are looking for a Cloud Principal Hardware Engineer to scale AFPD across hardware components in the Azure fleet and to lead hands-on hardware resiliency for the platform.
The opportunity
Every hour a server sits unhealthy is an hour of lost customer capacity. AFPD turns hardware, firmware and OS telemetry into early detection and prediction of failures, moves customer workloads off failing nodes, and sends a targeted repair for the part that failed. Repair outcomes and supplier failure analysis then improve the next rule or model. You will be working on incorporating into the platform components such as storage, memory, power supply, CPU, GPU, NIC, HBA/FPGA, cables and the rest of the server, and you will be the subject matter expert the team relies on when a failure needs to be prevented. Across millions of nodes, each improvement means fewer VM interruptions for customers and faster return of online capacity. Why this role: Your impact shows up in fleet availability and customer uptime, quarter over quarter. You will set the hardware direction for a platform that influences production infrastructure at a scale, and you will work directly with Azure engineering, component teams and leading hardware suppliers. #SCHIE #azurehwjobs Responsibilities Lead failure analysis across server components: characterize failure modes from fleet telemetry and logs and confirm root cause with repair results and supplier failure analysis. Define the signals, thresholds, targeted diagnostics and repair recipes, and work with component teams for the root cause analysis. Improve isolation accuracy so each repair address the source of failures: improve efficacy of the detection and repair services. Partner with software engineers and data scientists on detection rules and prediction models; set precision targets and validation gates before a rule or model takes action on production nodes. Use AI agents and machine learning to automate failure analysis, live-site triage and supplier RCA follow-up, and to shorten the time from a new failure signature to a detection rule. Lead root cause and corrective actions with server components, including firmware fixes, and feed the learnings back into the platform. Lead as the senior hardware escalation point for live-site incidents, from triage through root