Solutions Architect, LLM Model Builder

NVIDIA • Santa Clara, CA

7d • $152,000 - $241,500

About The Position

NVIDIA is seeking a Solutions Architect to join its Partner Solutions Architecture team, focused on enabling partners working with reasoning models, multimodal models, and production inference. The role combines strategic technical expertise with hands-on advising to help partners build, benchmark, fine-tune, optimize, and deploy foundation model solutions for customer workloads.

The team serves as a trusted advisor to NVIDIA’s ecosystem, translating customer requirements into architectures, benchmark recipes, cluster test plans, compute sizing, and production-readiness plans. The position supports partner and customer engagements across the full-stack accelerated computing platform and helps accelerate time to value.

Requirements

• MSc or PhD in Computer Science, Electrical Engineering, Software Engineering, ML Engineering, or a related field, or equivalent experience.
• At least 5 years of relevant experience working with LLMs, VLMs, and large-scale inference systems.
• Hands-on expertise in fine-tuning, benchmarking, evaluation, optimization, and production deployment, such as in a Research Engineer or Deep Learning Engineer role or equivalent.
• Strong understanding of foundation models across data preparation, fine-tuning, post-training, evaluation, and inference.
• Familiarity with reasoning models, reinforcement learning, and synthetic data generation and evaluation workflows.
• Strong programming skills in Python and hands-on experience with PyTorch, JAX, or TensorFlow.
• Familiarity with Nemotron, NeMo, Dynamo, TensorRT-LLM, Triton, vLLM, and similar inference and optimization stacks.
• Strong communication and presentation skills, with the ability to advise technical teams and executives.

Nice To Haves

• Experience helping partners or customers deploy large-scale AI systems in production.
• Experience building benchmark suites, fine-tuning recipes, sizing calculators, or TCO models for AI workloads.
• Strong knowledge of GPU infrastructure, including NVLink, InfiniBand, MPI, NCCL, or adjacent cluster technologies.
• Active open-source contributions in model tooling, inference, evaluation, or performance optimization.
• Comfort moving between deep technical reviews, architecture guidance, benchmarking, and partner enablement.

Responsibilities

• Serve as the lead technical advisor for partners delivering reasoning, multimodal, fine-tuning, and model-serving solutions.
• Guide partners in selecting approaches for customer workloads, including fine-tuning, distillation, quantization, compression, benchmarking, and evaluation.
• Define benchmark plans, synthetic data and evaluation workflows, and repeatable validation recipes.
• Advise on compute planning, including cluster sizing, GPU and network selection, storage, memory tradeoffs, latency and throughput targets, and production-readiness testing.
• Guide inference architecture decisions involving prefill and decode tradeoffs, batching, routing, disaggregated inference, and serving efficiency.
• Develop reference architectures, playbooks, benchmark recipes, TCO calculators, and sizing models using CUDA, NeMo, Nemotron, Dynamo, TensorRT-LLM, Triton, NIMs, and related tools.
• Support pre-sales and post-sales engagements by translating complex model and infrastructure topics for partner and customer teams.

Benefits

• Competitive salaries.
• Comprehensive benefits package.
• Equity.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior