PhD in Computer Architecture for Real-Time AI at the Edge Eindhoven University of Technology in Netherlands

Join the NWO Perspectief FIND program at Eindhoven University of Technology and contribute to the design of next-generation computer architectures for real-time, energy-efficient AI at the edge. This PhD position is part of a large, interdisciplinary research initiative involving five universities, eleven companies, and two knowledge institutes, targeting breakthroughs in running large AI models on embedded and edge systems under strict timing, energy, and memory constraints. The program is strongly industry-connected, with partners such as ASML, NXP, Canon, ASMPT, and others, and emphasizes edge deployment, privacy, and timely decision-making for high-tech industry applications.

Your research will focus on developing novel computer architectures and hardware-aware optimization techniques, including co-design with accelerators like FPGAs, GPUs, and near-memory systems. You will address the challenges of deploying foundation models (large AI models) in industrial edge environments, where current general-purpose CPUs and GPUs fall short in terms of latency, energy consumption, and memory footprint. The project involves joint optimization of algorithms and hardware, aligning architectural decisions (such as token merging, sparsity exploitation, and quantization) with hardware datapaths and memory hierarchies to achieve predictable real-time performance and energy efficiency.

Prototyping and validation will be conducted in collaboration with ASMPT, a global leader in electronics manufacturing solutions, ensuring that your research has direct industrial impact. The Electronic Systems (ES) group at TU/e, renowned for design automation and embedded systems research, will provide a vibrant academic environment with access to world-class facilities and support. You will be part of a diverse team of professors, postdocs, and PhD candidates, and benefit from interdisciplinary collaboration and high-quality training programs.

Applicants should have a master’s degree in Computer Engineering, Electrical Engineering, Embedded Systems, or a related field, with strong hardware design skills and expertise in computer architecture, digital design, real-time systems, and hardware/software co-design. Experience with hardware acceleration (FPGAs, GPUs, SoCs), low-power design, and deep learning frameworks (PyTorch, TensorFlow) is highly desirable. The position requires excellent English communication skills (C1 level) and motivation to develop teaching and coaching abilities.

Employment is full-time for four years, with a competitive salary (scale P: €3,059–€3,881/month), year-end bonus, vacation pay, pension scheme, paid parental leave, commuting and internet allowance, and a 30% tax compensation scheme for international candidates. The university offers a dynamic, international setting, excellent technical infrastructure, and support for personal and professional growth.

To apply, submit your application online with a cover letter, CV, and contact details for three references. The vacancy remains open until filled, with priority given to complete applications. For further information, contact Dr.ir. Sander Stuijk ([email protected]) or HR advisor Kevin Caris ([email protected]).