Senior GPU Kernel Engineer for LLM AI Workloads

AMD; San Jose (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

AMD is seeking a senior technical member to design, develop, and optimize high-performance GPU kernels for LLMs and ML models on AMD GPUs. You will analyze bottlenecks and craft architecture-aware optimizations to maximize compute efficiency and memory bandwidth.

You will collaborate with GPU architecture, compiler, library, and framework teams to deliver scalable solutions, provide technical leadership, and drive projects independently in a fast-paced environment.

Qualifications

• Strong experience in GPU programming, performance analysis and system software development.
• You will provide technical leadership, guiding teams and driving projects independently.
• Strong C++ and Python skills.
• Must be a self-starter, and able to independently drive tasks to completion.

Responsibilities

• Architect and Optimize AI Workloads: Lead the optimization of LLMs, quantized models, and emerging AI workloads for AMD GPU architectures.
• Develop High-Performance GPU Kernels: Design, implement, profile, and optimize GPU kernels, with a strong focus on maximizing hardware utilization and application performance.
• Drive Performance Analysis: Perform deep performance characterization across compute, memory bandwidth, cache hierarchy, synchronization, kernel scheduling, and execution efficiency to identify and eliminate bottlenecks.
• Advance Compiler and Code-Generation Technologies: Develop compiler-level optimizations and code-generation techniques using LLVM, MLIR, and related technologies to achieve portable and scalable GPU performance.
• Innovate with AI-Assisted Development: Build AI-powered coding agents and automation workflows for kernel generation, optimization, benchmarking, and validation.
• Collaborate with model and infrastructure teams to improve end-to-end AI system performance.

Skills

C++

Python

GPU programming

Performance analysis

Leadership

Analytical

Self-starter

Education

Bachelor's or Master's in computer engineering or computer science

Tools

HIP

CUDA

LLVM/MLIR

Job description

AMD is seeking a senior technical member to design, develop, and optimize high-performance GPU kernels for LLMs and ML models on AMD GPUs. You will analyze bottlenecks and craft architecture-aware optimizations to maximize compute efficiency and memory bandwidth.

You will collaborate with GPU architecture, compiler, library, and framework teams to deliver scalable solutions, provide technical leadership, and drive projects independently in a fast-paced environment.