Senior Research Engineer – Generative AI Inference

NVIDIA; Seattle (WA); On-site; USD 192,000 - 356,500; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA in Seattle is seeking a Senior Research Engineer focused on Generative AI inference. You will design routing policies for LLM traffic to balance accuracy and latency, and build agentic benchmarks to calibrate models across NVIDIA’s serving stack.

Collaborate with research and engineering teams to contribute design docs, code reviews, and open-source contributions, while mentoring junior engineers and driving experimental validation.

Qualifications

• Bachelor's or Master's degree in Computer Science or equivalent.
• 8+ years of industry experience in Deep Learning frameworks (PyTorch or TensorFlow).
• Experience designing or running LLM evaluations/benchmarks and drawing sound conclusions.
• Understanding of ML, DNNs, NLP, or Speech Recognition.
• Empirical research mindset: form hypotheses, calibrate, iterate.
• Strong communication and teamwork; mentoring junior engineers is a plus.
• Desire to grow and learn new things.
• Strong CS fundamentals: algorithms, data structures, parallel and distributed computing.

Responsibilities

• Design and evaluate routing policies for LLM traffic to optimize model usage.
• Build and run agentic benchmarks to measure algorithm quality and calibration data.
• Ship to an open-source repo with design docs, code reviews, and community contributions.
• Collaborate with engineering teams to integrate software across NVIDIA's serving stack.

Skills

Deep learning frameworks experience

LLM benchmarking

Machine learning fundamentals

Distributed & parallel computing

Algorithms & data structures

CUDA / GPU knowledge

Mentoring experience

Education

Bachelor's or Master's in Computer Science

Tools

CUDA

PyTorch

TensorFlow

Job description

NVIDIA in Seattle is seeking a Senior Research Engineer focused on Generative AI inference. You will design routing policies for LLM traffic to balance accuracy and latency, and build agentic benchmarks to calibrate models across NVIDIA’s serving stack.

Collaborate with research and engineering teams to contribute design docs, code reviews, and open-source contributions, while mentoring junior engineers and driving experimental validation.