Principal Software Engineer, At-Scale Reliability and Fleet Intelligence — CSP Engagements

## Principal Software Engineer, At-Scale Reliability and Fleet Intelligence — CSP Engagements Apply locations:
US, CA, Santa Claratime type:
Full time posted on:
Posted 2 Days Agojob requisition :
JR2020320

We're looking for a Principal Software Engineer to join our CSP Engagements team as the technical focal point for fleet-scale reliability, working directly with engineering teams of key CSP / hyperscale customers to ensure NVIDIA platforms achieve target MTBI (Mean Time Between Interruptions) in production. In this role, you will augment NVIDIA's internal software/firmware and quality teams with a dedicated CSP-facing focus.

You will drive work streams with CSP engineering teams to build shared understanding of reliability software/firmware architecture, methodology, incorporate their fleet telemetry and failure data into NVIDIA's improvement priorities, and validate that reliability improvements measured in the lab translate to real customer environments. Your cross-CSP visibility enables you to distinguish systemic architectural gaps from environmental or configuration-specific issues that no single customer engagement could identify alone.
** What you'll be doing:
*** Drive reliability work streams with CSP engineering teams — ensuring shared understanding of MTBI measurement methodology, failure classification, and health monitoring architecture
* Gather and synthesize CSP fleet reliability data — identify failure patterns that appear across multiple customers and champion improvements back into NVIDIA's firmware, driver, and hardware teams
* Define consistent MTBI measurement methodology that works across different CSP monitoring environments and operational practices
* Conduct fleet-scale failure pattern analysis using statistical methods (Pareto, survival analysis, Weibull) to classify failures as systemic, environmental, or configuration-specific
* Drive fleet health monitoring integration architecture — ensure NVIDIA's health agents, telemetry, and reporting align with CSP operational workflows and automation
* Define burn-in reliability test environment and cluster certification criteria in collaboration with quality teams, validating with customers that criteria are meaningful
* Collaborate with CSPs to ensure reliability-related integration work (health monitoring deployment, telemetry pipeline, alerting configuration) is complete ahead of at-scale launch
* Develop predictive failure models using fleet telemetry and validate their effectiveness in customer environments
** What we need to see:
*** 15+ years of experience in systems software at datacenter scale, or reliability engineering with focus on at-scale challenges.
* BS or MS in Computer Science, Electrical Engineering, Statistics, or related field (or equivalent experience)
* Deep expertise in multi-NUMA, rack-scale system software and firmware. Statistical failure analysis methods: MTBF/MTBI calculation, Pareto analysis, root cause classification
* Experience with fleet-level telemetry and observability systems: time-series databases, anomaly detection, health scoring, event correlation
* Understanding of hardware failure modes in large-scale GPU/accelerator deployments — ability to classify and prioritize across compute, interconnect, memory, power, and thermal domains
* Experience defining or operating burn-in, stress testing, or certification frameworks for complex hardware systems. Familiarity with predictive maintenance or anomaly detection approaches applied to fleet health data
* Customer obsession — genuine passion for understanding fleet reliability challenges at scale and translating them into actionable engineering priorities
* Strong communication — ability to present statistical reliability findings to both deep…