Senior AI Accelerator Architect - Edge & Data Center

QUALCOMM, Inc.; Austin (TX); On-site; USD 127,000 - 190,000; Full time

Job summary

Qualcomm Technologies, Inc. is seeking an AI Accelerator Architecture Engineer to drive functional, performance, and power improvements for on-device AI training and inference.

You will work on scalable accelerator architectures, memory subsystems, and NoC designs to support state-of-the-art large models, with a strong emphasis on power efficiency and pre-silicon performance modeling. The ideal candidate will have 3+ years in hardware engineering for GPUs/accelerators, deep knowledge of

Qualifications

• Master's degree in CS/Engineering/IS or related field.
• 3+ years hardware engineering experience in GPUs/accelerators for AI training.
• In-depth knowledge of NVIDIA/AMD GPUs and architectures.
• Knowledge of LLM architectures and HW requirements.

Responsibilities

• Understand ML network design trends and translate to SW/HW requirements.
• Define hardware requirements for AI training systems with customers.
• Analyze current accelerator and GPU architectures and propose improvements.
• Architect enhancements for efficient AI model training.
• Design flexible computational blocks for diverse data types and precisions.

Skills

C++

Python

GPU architectures

Machine learning

Communication

Education

Master's degree in CS/Engineering/IS

Bachelor's degree in CS/Engineering/IS

PhD in CS/Engineering/IS

Tools

DeepSpeed

TensorFlow

PyTorch

NVIDIA DGX

NoC/SoC design

Job description

Qualcomm Technologies, Inc. is seeking an AI Accelerator Architecture Engineer to drive functional, performance, and power improvements for on-device AI training and inference.

You will work on scalable accelerator architectures, memory subsystems, and NoC designs to support state-of-the-art large models, with a strong emphasis on power efficiency and pre-silicon performance modeling. The ideal candidate will have 3+ years in hardware engineering for GPUs/accelerators, deep knowledge of