PhD in Computer Architecture for Real-Time AI at the Edge

# PhD Position: Computer Architectures for Edge AI at TU/e

## Overview
Join the NWO Perspectief FIND program to design next-generation computer architectures for running large AI models on embedded and edge systems under strict timing, energy, and memory constraints. This research focuses on hardware-aware optimization and co-design with accelerators (FPGAs, GPUs, near-memory systems) to achieve real-time, energy-efficient AI for high-tech industry applications.

## Research Focus
The PhD position focuses on:

• Design of novel computer architectures enabling large AI models to run on embedded and edge systems with strict timing, energy, and memory constraints
• Model compression and co-optimization with hardware accelerators
• Joint optimization of algorithms and hardware to ensure architectural decisions are aligned with hardware datapaths and memory hierarchies
• Exploration of token merging, sparsity exploitation, and quantization techniques
• Prototyping and validation in collaboration with ASMPT, a leading semiconductor manufacturing company

## Industrial Applications
Target domains include semiconductor manufacturing, printing systems, automotive radar, smart mobility cameras, and HealthTech. The research aims to enable on-device AI with low latency, privacy protection, and resilience.

## Work Environment

• Embedded in the Electronic Systems (ES) group at Eindhoven University of Technology (TU/e)
• Close collaboration with ASMPT and other industry partners
• Part of a larger FIND program involving 5 universities, 11 companies, and 2 knowledge institutes
• International, interdisciplinary research setting

## Teaching Component
Minimum 10% of four-year employment on teaching tasks (maximum 15% per year).