Latest AI JobsBrowse Jobs
snowflake logo

Senior/Staff System Research Engineer – LLM Inference Optimization

snowflakeUS-WA-BellevueEngineering
Apply on company site
Salary
$236K – $330K
Job type
Full-time
Work mode
Remote
Experience
Senior · 5+ yrs
Posted
21 Sept 2026
Status
Open · checked 7d ago

About this role

The Senior/Staff System Research Engineer – LLM Inference Optimization role is based in Bellevue, Washington at Snowflake and suits experienced engineers focused on large language model inference and distributed systems. This position involves advancing high-performance inference technology and shaping production-scale artificial intelligence operations.

Role Overview

Snowflake seeks systems developers and researchers for the AI Research team in Bellevue, Washington. The position focuses on building the next generation of intelligent and high-performance inference systems across the full stack, ranging from distributed serving to GPU kernels.

Key Responsibilities

Design and develop high-performance large language model inference systems covering distributed serving, runtime systems, GPU execution, and performance-critical kernels.
Create novel techniques to enhance generation speed, inference latency, throughput, memory efficiency, scalability, and cost reduction.
Research advanced methodologies including speculative and parallel decoding, prefill and decode disaggregation, continuous batching, KV-cache management, quantization, and adaptive parallelism.
Build adaptive inference systems that automatically optimize execution for new hardware platforms, workloads, and model architectures.
Apply artificial intelligence-driven approaches to systems engineering for automated profiling, bottleneck identification, debugging, and performance tuning.
Formulate hypotheses, prototype solutions, and drive research concepts into production scale.
Establish distributed inference strategies across multiple GPUs and nodes utilizing tensor, sequence, pipeline, data, and expert parallelism.
Optimize GPU kernels and operators for attention mechanisms, mixture-of-experts, and communication components.
Profile end-to-end workloads and publish innovations through technical blogs and academic conferences.

Requirements

Bachelor’s degree in Computer Science, Electrical Engineering, or a related field.
Minimum of 5 years of professional experience in large language model inference systems, distributed artificial intelligence systems, GPU systems, or high-performance computing.
Familiarity with modern serving frameworks such as vLLM, SGLang, or TensorRT-LLM.
Hands-on background in optimizing inference runtimes, scheduling, batching, and KV-cache management.
Understanding of GPU architectures with practical coding capability in CUDA, Triton, CUTLASS, cuBLAS, or cuDNN.
Proficiency in profiling system performance using tools like Nsight Systems or Nsight Compute.

Nice to Have

Master's degree or PhD in a related field.
Experience applying artificial intelligence-native engineering practices to accelerate software development, experimentation, debugging, or system adaptation.

Skills and Tools

vLLM
SGLang
TensorRT-LLM
CUDA
Triton
CUTLASS
cuBLAS
cuDNN
Nsight Systems
Nsight Compute
Distributed serving systems
GPU execution kernels

Salary and Benefits

The salary range is $236K to $330K. Detailed benefits information is available on the official company careers portal.

About the Company

Snowflake operates globally to power the agentic enterprise, encouraging experimental mindsets and low-ego collaboration among team members.

How to Apply

Interested candidates should apply on the official company careers page at careers.snowflake.com.

Skills and tech stack

vLLMSGLangTensorRT-LLMCUDATritonCUTLASScuBLAScuDNNNsight SystemsNsight ComputeTensorFlowLLMMachine Learning

Questions candidates ask

What is the location of the Senior/Staff System Research Engineer role?

The position is located in Bellevue, Washington, within the United States.

What is the salary range for this position?

The salary range is between $236K and $330K.

How many years of experience are required?

Candidates must have 5 or more years of experience in large language model inference systems, distributed artificial intelligence systems, GPU systems, or high-performance computing.

What educational qualifications are required?

A Bachelor's degree in Computer Science, Electrical Engineering, or a related field is required, while a Master's degree or PhD is preferred.

Where should I submit my job application?

Applicants should submit their applications directly through the official company careers page at careers.snowflake.com.

Ready to apply?

The application is on the official snowflake careers page.

Apply on company site