Hybrid Principal AI Engineer - Generative Models at Scale

AMD; San Jose (CA); Hybrid; USD 250,000 - 360,000; Full time

Job summary

AMD in San Jose, CA is seeking a Principal Engineer to advance Generative AI training and inference at scale. You will join the AI Models and Applications team to design efficient transformer architectures and scalable distributed training on AMD hardware.

Qualifications include a PhD or master's in CS/EE/math, deep experience with LLMs or image/video models, and fluency with PyTorch, JAX, vLLM, SGLang, or MuJoCo.

Qualifications

• You have a deep technical understanding of Generative AI applications (LLMs, 3D World/Action models, or image/video generation).
• You have experience training models at scale and enabling distributed training and inference on AMD devices.
• You hold a PhD or master's in a relevant field and have several years in AI/deep learning.

Responsibilities

• Propose and apply innovative techniques for both training and inference, including transformer architectures and parallelism strategies.
• Develop novel architectures for Generative AI models and showcase benefits on AMD platforms.
• Collaborate with open-source communities to integrate AMD-optimised models and publish training recipes.
• Co‑optimise end‑to‑end performance with software and hardware teams; promote scalable AI workflows.

Skills

Generative AI expertise

Distributed training

Technical leadership

Education

PhD or master's in CS/EE/Math

Tools

PyTorch

JAX

vLLM

SGLang

MuJoCo

Job description

AMD in San Jose, CA is seeking a Principal Engineer to advance Generative AI training and inference at scale. You will join the AI Models and Applications team to design efficient transformer architectures and scalable distributed training on AMD hardware.

Qualifications include a PhD or master's in CS/EE/math, deep experience with LLMs or image/video models, and fluency with PyTorch, JAX, vLLM, SGLang, or MuJoCo.