Top Tech Jobs & Startup Jobs in Seattle, WA

Advise ISVs on designing, deploying, and optimizing large-scale accelerated AI infrastructure. Lead architecture reviews, POCs, benchmarks, and production deployment guidance across compute, networking, storage, containers, orchestration, observability, and CI/CD. Create reference architectures, sizing guidance, technical playbooks, demos, and whitepapers. Support cluster monitoring, reliability, and performance improvements. Up to 20% travel for customer engagements.

Develop and maintain AI networking communication frameworks, production benchmarking tools, and automation for large-scale AI training/inference. Enable models in benchmarking infrastructure, analyze end-to-end performance, perform large-scale parameter searches, and collaborate with networking and hardware teams to co-design features and optimize system performance.

Develop and implement compiler optimization algorithms for deep learning workloads, analyze networks, debug and profile GPU performance (CUDA), define public APIs, and deliver ahead-of-time and just-in-time compilation techniques for NVIDIA GPUs to improve inference performance, build times, and memory efficiency.

The role involves deploying AI networking solutions in data centers, driving architectural decisions, troubleshooting network issues, and collaborating on customer projects for NVIDIA's technology solutions.

Design, build, and optimize GPU‑accelerated inference software and open‑source frameworks (vLLM, SGLang, FlashInfer). Improve performance and scale for LLMs and generative models across NVIDIA GPUs and edge SoCs using CUDA, CUTLASS, Triton, NCCL and profiling tools. Collaborate across teams to enable efficient model serving and deployment.

Design and evaluate routing policies for LLM traffic, build and run agentic benchmarks, produce calibration data and routing profiles, contribute code and docs to open-source, and collaborate across engineering teams to integrate optimized inference software into NVIDIA's serving stack.

Design and develop scalable infrastructure for deep learning libraries, automate build/test processes, and contribute to open-source tools.

The role involves performing data analysis for datacenter applications, using machine learning for GPU systems, and providing insights on trends to enhance workflows.

Profile, benchmark, and optimize large-scale LLM training and inference across GPUs, CPUs, HCAs, and switches. Focus on collective communication and high-performance networking, build PyTorch trace-based profiling and replay tools, define performance test plans, and collaborate across hardware and software teams to identify bottlenecks and achieve performance targets.

Engage industry enterprise developers and AI startups to accelerate adoption of NVIDIA GPU-accelerated platforms. Provide technical guidance, create developer-facing content (samples, tutorials, blogs, trainings), collaborate with internal SDK and research teams, troubleshoot implementation issues, and contribute to hands-on training and DLI courses as an instructor.