Build advanced AI models on real EEG data and contribute to the next generation of neurotechnology.
Regulations are currently being finalized
An initiative by
Powered by
Real EEG datasets
What is it
“Collaborating with Université Paris-Saclay was a natural step for HABS. It brings together academic excellence and real-world neuro-AI challenges.”
The HABS Neuro-AI Hackathon brings together Master’s students (M1 & M2) and engineering students from their second year onwards in artificial intelligence, data science and neuroscience to tackle real challenges in brain signal analysis.
Participants will work with EEG datasets provided by HABS and develop machine learning approaches to improve real neuro-AI pipelines. Over four weeks, teams will explore innovative methods in signal preprocessing, multimodal AI and real-time deployment while receiving mentorship from the HABS scientific team.
The hackathon aims to generate new ideas, build meaningful prototypes and connect talented students with real research challenges, while offering potential internship and recruitment opportunities with HABS for top-performing participants.
The challenge
Participants will work with real neurophysiological datasets provided by HABS and develop machine learning approaches to improve brain signal analysis. The challenge focuses on advancing neuro-AI pipelines, from signal preprocessing to intelligent model deployment.
Teams will work across the full neuro-AI pipeline: signal preprocessing, feature extraction, model development and evaluation. The objective is to design robust approaches that can contribute to future brain-computer technologies.
Students will explore real-world neuro-AI problems using EEG datasets and advanced machine learning techniques. The goal is to experiment, prototype and develop innovative solutions that could improve HABS research pipelines.
Tracks
Data Preprocessing
Work with raw physiological data and learn to clean, filter and structure biosignals (EEG, ECG, respiratory rate, ...) for machine learning models.
Multimodal AI
Explore how EEG signals can be combined with other data to improve AI models and better understand cognitive states.
Real-time Deployment
Design pipelines capable of processing brain signals in real time and moving from experimental models to concrete applications.
Calendar
September 11, 2026
Optional first meeting at Saclay
Kick-off talk — meet the HABS team
November 2, 2026
HACKATHON starts + mandatory kick-off
Team formation & launch
November, 2026
Development phase
Work with HABS mentors
November 30, 2026
Final presentations
Awards ceremony
Podium
Teams will compete to develop innovative solutions using real EEG data.
The three best teams will be rewarded for their creativity, technical excellence, and impact.
The prize will be awarded to each member of the winning team.
The winning team
Together with
Benefits
A Professional & Hands-on Experience
Work with large-scale EEG datasets (up to 1 petabyte of data) to clean, structure and analyze real brain signals for AI applications.
Work alongside professionals and receive guidance from the HABS scientific and technical teams.
Top-performing participants will access internship and recruitment opportunities with HABS.
Output
Documented Code
Teams will submit a clean and well-documented notebook or code repository.
Final Presentation
Each team will present their approach and results to the final jury.
Each track will be evaluated using a dedicated evaluation grid. Projects will be assessed based on technical quality, relevance of results, teamwork, and clarity of the final deliverable, including the final presentation on November 30.
Discover HABS in action
Regulations are currently being finalized.