Top Tech Jobs & Startup Jobs

The role involves benchmarking deep learning models on NVIDIA GPUs, improving software performance, and developing automation tools for performance optimization.

Lead productization of NVIDIA HGX GPU boards for datacenter servers: define and implement manufacturing test specs, debug complex HW/FW/SW interactions, develop diagnostic tests, ensure DFx and signal integrity, optimize production lines and yield, collaborate with contract manufacturers and quality teams, and drive continuous improvement to meet cost, quality, and DPPM targets.

Lead go-to-market planning and content for NVIDIA AI on cloud platforms. Create customer success stories, manage events, align partner messaging, integrate models/software into marketing, and support partner and campaign teams. Travel to trade shows and trainings as needed.

Join NVIDIA's Public Sector DevTech team to GPU-accelerate applications, optimize parallel algorithms and data structures, develop reference code and contribute across the software stack, collaborate with architecture and tooling teams, and engage with external developers. Occasional travel for conferences and on-site visits.

Stabilize and optimize customer AI/HPC clusters post-deployment, address networking change requests, validate and troubleshoot InfiniBand/Ethernet fabrics, collaborate with CBU DevOps and Infrastructure Solutions Architects, and support pre- and post-sales networking engagements to help customers expand clusters.

Perform physical design and timing analysis for GPUs, CPUs, DPUs and SoCs across block, cluster, and full-chip levels. Drive frontend/backend implementation including synthesis, equivalence checking, floorplanning, timing constraints, STA-based timing convergence, power convergence, and ECO implementation while collaborating cross-functionally.

Design, develop, and optimize high-performance C++ systems for ultra-low-latency media streaming. Build and improve streaming pipelines (WebRTC, GStreamer, RTP/RTCP), analyze performance across server, network, and media pipelines, and collaborate with encoding, cloud networking, and systems teams to define KPIs and reduce latency for cloud gaming and related use cases.

Build and optimize quantized and sparse inference recipes for LLMs, implement performant kernels and model-level changes in inference engines, own model export pipelines, develop benchmarking and numerics-debugging tools, improve developer productivity (CI/build/training), and collaborate with partner teams to productize across Megatron-LM, ModelOpt, and vLLM.

The Technical Program Manager will lead NPI programs, coordinating with engineering and cloud service partners to enhance infrastructure and automate workflows.

Design, develop, and debug system software for NVIDIA Halos across QNX, Linux, and hypervisors. Build BSP, boot, drivers, device management, networking, diagnostics, and I/O virtualization for NVIDIA SoCs. Ensure safety/security per ISO 26262/21434, use formal methods, optimize platform bring-up and performance, integrate Ethernet switching and automotive networking (PTP/TSN), and use AI-assisted engineering tools to accelerate development.