Postdoc Position: Efficient foundation model inference across the computing continuum Aarhus University in Denmark

Aarhus University is offering a 1-year postdoctoral position, with the possibility of a 1–2 year extension, in efficient foundation model inference across the computing continuum. The position is based at the Department of Electrical and Computer Engineering and is centered on communication-efficient and distributed inference from cloud to edge, with a strong focus on scalable, low-latency, resource-aware deployment of large foundation models.

The successful candidate will contribute to research topics such as token compression and adaptive token pruning, distributed and collaborative inference strategies, mixture-of-experts architectures for scalable inference, resource-aware and latency-constrained optimization, and edge intelligence/on-device foundation model deployment. The work is embedded in a dynamic and collaborative environment with opportunities to publish in top-tier venues and engage with academic and industry partners within a Horizon Europe project.

The ideal applicant should hold a Ph.D. in Computer Science, Computer Engineering, Electrical Engineering, or a related field. A strong background in deep learning is required, particularly in transformers and foundation models, together with strong Python programming skills and experience with frameworks such as PyTorch. Experience with distributed systems and edge AI is also expected, along with an excellent publication record relative to career stage and strong English communication skills. Experience with stream data, goal-oriented communications, and collaborative cross-cultural research is considered an advantage.

The position is supervised by Professor Qi Zhang, whose research interests include edge intelligence, goal-oriented semantic communications, Internet of Things, and analytics on compressed data. The postdoc will join a research environment at Aarhus University that emphasizes high-impact research, diversity, and strong support for career development.

The appointment is full-time and fixed-term, starting on September 1, 2026 or as soon as possible thereafter, with an initial end date of August 31, 2027 (or 12 months from the actual start date). An extension of 1–2 additional years may be offered subject to satisfactory performance. The application deadline is 2026-04-30. Applicants must submit materials in English through Aarhus University’s recruitment system.

This opportunity is especially relevant for researchers in deep learning systems, distributed AI, edge intelligence, and efficient inference for large-scale foundation models.