Eindhoven University of Technology

Decentralized AI Research Lab

We study how models can learn from large amounts of unlabelled real-world data such as audio, biosignals, sensor streams and text, and keep working once that data is spread across sites and devices rather than sitting in one place.

Directions

Multimodal & foundation models

Pretraining and post-training that align signals, audio and language in a shared representation, so downstream tasks need few labels or none.

Language models & agentic systems

Agents that plan, write code, use tools and critique each other, including systems that design and run machine learning pipelines on their own.

Federated & efficient learning

Training across many clients under noisy labels, heterogeneous sensors and tight communication budgets, and inference small enough for embedded hardware.

People

PhD researchers

The group is led by Aaqib Saeed.

Jialu TangSignal and language models, zero-shot inference Haoyuan LiFederated learning, LLM agents for ML systems Tsai-Ning WangAudio understanding, post-training alignment Ran PiaoCross-modal representation learning, audio language models Mustafa IlerisoySafety of multimodal audio language models Hamza ShafiqueFoundation multimodal biosignal models Yishan WangAgentic AI, multimodal learning

Joining

Openings are announced through TU/e and on LinkedIn. MSc students looking for a thesis project, and researchers interested in a visit or a joint proposal, are welcome to write. Please include something you have built.