Principal Software Engineer, Profiling Services

NVIDIA•Austin, CA

About The Position

Design and ship an Always‑On, low‑overhead GPU profiling service that runs in production, scales across cluster environments, and delivers actionable insights for ML workloads. You will lead the architecture and hands‑on delivery across system software , drivers , and CUDA to make profiling continuously available and reliable.

Requirements

BS or MS degree or equivalent experience in Computer Engineering, Computer Science, or related degree.
15+ years of system‑level C/C++ development, including concurrency, memory management, and performance engineering.
Expertise with profiling/tracing stacks for CPU/GPU (e.g., CUPTI, Nsight, performance counters, event correlation) and debugging concurrent systems.
Deep hands‑on CUDA and GPU architecture knowledge (runtime/driver APIs, CUDA streams/graphs, kernel behavior).
Proven experience designing and shipping production quality system software or drivers with strict reliability, observability, and performance constraints.
Demonstrated technical leadership : defining architecture and success metrics, and translating abstract product visions into actionable technical roadmaps with fast-paced, multidisciplinary teams.
Strong interpersonal, verbal, and written communication; able to influence across organizations and build trust with external collaborators.

Nice To Haves

Track record building continuous/always‑on or multi‑client profiling systems with predictable overhead at scale.
Hands-on experience tuning ML training/inference loops based on deep profiling analysis.
Familiarity with ML ecosystems (e.g., PyTorch , JAX ) and correlating application‑level events with GPU traces/metrics.
Strong background in translating profiling data into actionable performance insights (compute vs memory bound, bottleneck triage).
Experience with user‑mode driver development and integration with platform permissions/securityA models.

Responsibilities

Own the architecture for an Always‑On profiling service, defining interfaces, data flows, and scalability guarantees for multi‑process/GPU/node systems.
Drive low‑overhead, high‑reliability implementations in C/C++ , including IPC/shared memory, lock‑free buffers, and bounded CPU/memory budgets with clear benchmarks.
Lead end‑to-end feature delivery spanning user‑mode components, driver/platform layers, and performance counter/trace providers.
Establish profiling models that integrate with existing ML/AI workflows (e.g., PyTorch/XLA) to turn low‑level signals into actionable insights.
Set technical direction for an engineering team; mentor engineers, drive technical planning to mitigate architectural risks, and align roadmaps across internal and external partners.