Senior Software Architect, AI Systems and Networking

NVIDIA•Austin, TX

2d•$224,000 - $431,250

About The Position

NVIDIA’s Networking Systems & Software Architecture group includes an applied research team addressing infrastructure challenges in AI. The team develops systems-level software to move data between GPUs, nodes, and storage, working across low-level transport optimization, hardware-software co-design, and communication frameworks that integrate with production AI stacks. Its charter also includes emerging areas such as quantum computing interconnects. The Senior Architect owns modules and projects end-to-end, from scoping research questions through delivering production code. The role requires recognized technical expertise, leadership of technical decisions, drawing on research and industry ideas, and prototyping approaches to evaluate them. The work spans applied research and production engineering.

Requirements

12+ years in systems software and/or networking, with demonstrated ownership of complex projects.
MS, PhD, or equivalent experience in Computer Science, Computer Engineering, Electrical Engineering, or a related field.
Solid understanding of high-performance networking, including InfiniBand, RoCE, RDMA, NVLink, and GPUDirect.
Strong systems programming skills in C, C++, or Rust, with comfort in performance profiling and low-level debugging.
Understanding of ML systems concepts such as transformer architectures, KV cache mechanics, model parallelism, or distributed training and inference patterns.

Nice To Haves

Knowledge of ML inference frameworks such as vLLM, SGLang, and TensorRT-LLM, and their communication requirements.
Knowledge of storage networking, including NVMe-oF, GPUDirect Storage, and S3.
Background in Reinforcement Learning systems.

Responsibilities

Architect and implement high-performance communication and memory management libraries for distributed AI.
Drive hardware-software co-optimization with GPU, DPU, NIC, and switch teams using GPUDirect RDMA, NVLink, and next-generation interconnects.
Profile and optimize data movement across GPU memory, system DRAM, NVMe, and network fabrics.
Integrate networking capabilities into AI serving stacks such as vLLM, SGLang, and TensorRT-LLM.
Contribute to and maintain open-source projects.
Mentor engineers and conduct design reviews.
Prototype experimental technologies to evaluate their viability.
Own modules and projects end-to-end, from scoping research questions to shipping production code.