Postdoc Position: Efficient foundation model inference across the computing continuum (re-adverti...

Aarhus University (AU); Denmark; On-site; DKK 500,000 - 800,000; Full time

Benefits offered by this job

Access to state-of-the-art facilities

Support for research career development

Commitment to diversity, equity, and inclusion

Healthy work-life balance

Job summary

Aarhus University (AU) is seeking a motivated postdoctoral researcher for a 1-year position in efficient foundation model inference with extension possibilities. The successful candidate will engage in cutting-edge research, focusing on topics such as token compression, distributed inference, and edge intelligence.

This role requires a Ph.D. in a relevant field, a strong background in Deep Learning, and excellent programming skills. The work location will be at Helsingforsgade 10, Aarhus. Benefits include access to state-of-the-art facilities and support for research career development.

Qualifications

• Ph.D. in Computer Science, Computer Engineering, Electrical Engineering, or related field.
• Strong background in Deep Learning.
• Strong programming skills in Python and experience with deep learning frameworks.
• Experience with distributed systems and edge AI.
• Strong publication record in reputable conferences or journals.
• Excellent written and spoken English communication skills.

Responsibilities

• Contribute to research on token compression and adaptive token pruning.
• Research on distributed and collaborative inference strategies.
• Work on Mixture-of-Experts (MoE) architectures for scalable inference.
• Optimize resource-aware and latency-constrained inference.
• Engage in edge intelligence and on-device foundation model deployment.

Skills

Deep Learning (Transformers, foundation models)

Programming skills in Python

Experience with deep learning frameworks (e.g., PyTorch)

Experience with distributed systems and edge AI

Strong publication record

Excellent written and spoken English

Education

Ph.D. in Computer Science or related field

Job description

We are seeking a highly motivated and talented postdoctoral researcher for a 1‑year position (with the possibility of 1‑2 year extension) in efficient foundation model inference.

Expected start date and duration of employment

This is a full‑time, fixed‑term position for one year, starting on November 1, 2026, or as soon as possible thereafter.

Job description

The successful candidate will contribute to cutting‑edge research on topics including:

• Token compression and adaptive token pruning
• Distributed and collaborative inference strategies
• Mixture‑of‑Experts (MoE) architectures for scalable inference
• Resource‑aware and latency‑constrained inference optimization
• Edge intelligence and on‑device foundation model deployment

Your profile

Required qualifications:

• Ph.D. in Computer Science, Computer Engineering, Electrical Engineering, or a related field
• Strong background in Deep Learning (e.g., Transformers, foundation models)
• Strong programming skills in Python and experience with deep learning frameworks (e.g., PyTorch)
• Experience with distributed systems and edge AI
• Strong publication record in reputable conferences or journals relative to career stage
• Excellent written and spoken English communication skills

Additional qualifications that will be considered an advantage:

• Experience in working with stream data
• Experience in working with goal‑oriented communications
• Experience with collaborative, cross‑cultural research environments

What we offer

• Access to state‑of‑the‑art facilities and computing infrastructure
• Strong support for research career development, including mentoring and international networking opportunities
• A commitment to diversity, equity, and inclusion in all aspects of our work
• A workplace characterised by professionalism, equality, and a healthy work‑life balance

Place of work and area of employment

The place of work is Helsingforsgade 10, 8200 Aarhus N, and the area of employment is Aarhus University with related departments.

Deadline

Applications must be submitted by 30 August 2026.

Contact information

Professor Qi Zhang, [phone], [e-mail].

Legal statement

We welcome all applicants.