NVIDIA is looking for a technical team lead to run a small, high-impact team of software verification engineers. The team owns functional and performance verification of high-profile features in the DOCA SDK. These features let GPUs and SmartNICs move data over the network directly, with no host CPU involved. They power AI, cloud and accelerated-computing data centers on BlueField DPUs and ConnectX SuperNICs.
This is a technical leadership role. You will define the verification strategy and take technical ownership of your feature areas, including design reviews, root-cause analysis of complex failures, and technical discussions with architects and SDK developers. You will combine strong engineering judgment with the ability to develop and guide a high-performing team.
Why you’ll be doing:
- Lead, mentor and grow a small team of verification engineers, and take hands-on technical ownership of the team's feature areas.
- Define test plans and verification strategy for new features from the design stage. Cover API contracts, capability discovery, negative and boundary paths, resource lifecycle and clean teardown.
- Drive functional verification (C/C++ and CUDA test suites, integration and system tests) for GPU-initiated networking, RDMA-style data movement and high-rate Ethernet packet paths.
- Drive performance verification: bandwidth, latency, scaling and regression tracking. Build repeatable benchmarks and interpret the results.
- Debug across the full stack (GPU, NIC, host, DPU, drivers and firmware).
- Lead technical discussions with SDK developers, architects and product management. Push back on design and testability and sign off on release readiness.
- leverage AI tools in verification process to speed up test development and failure triage.
What we need to see:
- BS in EE, CS, CE or equivalent experience, with and strong C/C++ and Linux. Experience writing production-quality test code is a must.
- 8+ overall years in software development or verification and 2+ years as a team lead, tech lead or manager of engineers, while staying technically hands-on
- Hands-on networking experience in RDMA / InfiniBand / RoCE (queue pairs, completion queues, memory registration) or in the Ethernet datapath (RX/TX queues, offloads, DPDK or similar).
- Experience with verification methodology and performance testing .
- Strong debugging and communication skills.
Ways to stand out from the crowd:
- GPU and CUDA experience: writing or debugging CUDA kernels, persistent kernels and CPU-GPU synchronization, GPUDirect RDMA, and GPU-initiated communication.
- System-level performance tuning: NUMA, PCIe topology, hugepages, IOMMU/ACS and NIC tuning, and their effect on GPU-to-NIC data paths.
- Familiarity with the NVIDIA stack: DOCA SDK, BlueField DPUs, ConnectX NICs, rdma-core/ibverbs, DPDK, or packet steering and offload frameworks.
- Experience with firmware, driver or hardware-adjacent verification.
- Experience with Git/Gerrit, Jenkins and CI at scale, and with AI-assisted development and test generation.
NVIDIA is widely considered to be one of the technology world’s most desirable employers. We have some of the most forward-thinking and hardworking people in the world working for us. Are you a creative and autonomous engineer who loves a challenge? Come join our FW Verification team and help us build future interconnect architectures that will continue to drive us forward in the fields of High-Performance Computing, Graphics, AI and Networking.
We are an equal opportunity employer and value diversity at our company. We do not discriminate on the basis of race, religion, color, national origin, gender, sexual orientation, age, marital status, veteran status, or disability status. We will ensure that individuals with disabilities are provided reasonable accommodation to participate in the job application or interview process, to perform essential job functions, and to receive other benefits and privileges of employment. Please contact us to request accommodation.
