GenAI Inference Engineer (Forward-Deployed, AMD GPUs)

AMD; Medina; On-site; SAR 600,000 - 900,000; Full time

Job summary

AMD is seeking a Senior Member of Technical Staff for its Forward Deployed Engineering team in Medina. You will work directly with strategic customers to enable, optimize, and deploy GenAI inference workloads on AMD Instinct GPUs and CPUs.

You will bridge AI models with AMD hardware to deliver competitive performance across the inference stack. You will develop and tune GPU kernels, optimize pipelines, and collaborate with core teams to achieve targets while contributing to open-source projects

Qualifications

• 8+ years of relevant experience in GPU performance optimization
• Excellent knowledge of GPU architectures and programming models
• Hands-on experience with AMD ROCm (or CUDA) and kernel development
• Strong proficiency in Python and C++, with kernel development tools
• Experience with LLM inference frameworks and serving techniques

Responsibilities

• Partner directly with customers to enable and optimize LLMs and multimodal models on AMD Instinct GPUs
• Optimize GenAI inference pipelines and provide performance improvements
• Develop and tune low-level GPU kernels and tensor operations
• Profile workloads and identify bottlenecks across GPU-CPU boundaries
• Contribute optimizations to open-source AI frameworks and internal libraries

Skills

GPU architecture

Python

C++

Kernel development

Performance optimization

MoE architectures

Transformer models

Large language models

Education

BS, MS, or PhD in Computer Science or Electrical Engineering

Tools

Triton

HIP

CUDA

rocProfiler

ROCdebug-agent

Job description

AMD is seeking a Senior Member of Technical Staff for its Forward Deployed Engineering team in Medina. You will work directly with strategic customers to enable, optimize, and deploy GenAI inference workloads on AMD Instinct GPUs and CPUs.

You will bridge AI models with AMD hardware to deliver competitive performance across the inference stack. You will develop and tune GPU kernels, optimize pipelines, and collaborate with core teams to achieve targets while contributing to open-source projects