Software Architect, Network System Validation

NVIDIA is looking for a Software Architect to join the Network System Validation group. You will develop systems for validating advanced networking solutions across NVIDIA's complex AI cluster environments.

The Networking Business Unit (NBU) builds high-speed interconnects that switch thousands of GPUs into a single AI supercomputer.

As a Software Architect, you will:

• Design and implement network validation methodologies for system-level testing
• Build systems that orchestrate millions of concurrent IO operations, inject chaos at the infrastructure layer, and expose system faults
• Develop systems that capture, store, and process TBs of telemetry data for root-cause analysis and system-level insights
• Advance AI-driven approaches to test automation
• Establish engineering practices and produce clear, data-driven reports

Requirements:

• B.Sc. / B.A. in Computer Science, Electrical Engineering, or equivalent experience
• 12+ years of experience in developing large-scale software projects
• Strong distributed system experience and system-level debugging skills
• Strong software skills: C/C++/Rust, Python, Bash
• Experience building automation frameworks and validation tools
• Experience building large-scale infrastructure platforms or reliability engineering systems
• Deep understanding of distributed systems: concurrency, consistency models, fault tolerance, and large-scale system behavior under stress

Preferred qualifications:

• Background with RDMA / RoCE / AI networking (NCCL)
• Experience with L2/L3/L4 networking protocols
• Familiarity with NVIDIA networking solutions (ConnectX, SpecX, BlueField)
• Background in performance analysis, Kubernetes, or cloud environments
• Background in chaos engineering, fault injection, or simulation systems