Who is Bruno?
I am a Research Scientist at Yneuro (France 🇫🇷) and (Honorary) Research Associate at the University of California, San Diego (USA 🇺🇸). I obtained my PhD in Computer Science from Paris-Saclay University (France 🇫🇷) and Federal University of ABC (Brazil 🇧🇷), advised by Professors Sylvain Chevallier, Marie-Constance Corsi and Raphael Y. de Camargo.
Research Interests
My current research interests include Learning Representation from the time series (Decoding, Generating and Transferring knowledge), Brain-Computer Interfaces, Machine Learning for brain decoding, Benchmark, and Riemannian Geometry (mostly Symmetric Positive Definite Neural Networks - SPDNets).
Open Source Projects
I advocate for open source as the basis of reproducible science, while occasionally working with closed code. I lead the widely used Python libraries Braindecode and MOABB, and I collaborate with related open-source projects like MNE-Python, MONAI, MONAI Generative, and SpeechBrain.
From libraries to open datasets — a few of the things I build, organize, and maintain (full activity on GitHub):
Libraries & infrastructure
Mother of All BCI Benchmarks: reproducible evaluation of BCI pipelines across open datasets.
Geometric (Riemannian/SPD) deep learning library for neural decoding through trivialization.
Open data, tools, and compute resource for machine learning on neuroelectromagnetic data.
Community Involvement
During my PhD, I collaborated with research groups across the US (San Diego, San Francisco, Washington), UK, Ireland, Germany, Italy, Netherlands, Canada (Waterloo, MILA), Brazil (Sao Paulo), and France. Those collaborations turned into more than 16 publications (full/short papers, reports, abstracts), all listed on Google Scholar. I particularly enjoyed the self-contained, code-oriented projects 🧠⚙️. For academic cooperation, contact me via email or LinkedIn.
I organized the Braindecode Code-Sprint in the European summer of 2023, co-organized the workshop Designing Brain-Computer Interfaces from Theory to Real-Life Scenarios at the Graz BCI 2024 conference, and led the Special Session on Decoding the Brain Time Series at IEEE MLSP 2025.
I review for machine learning conferences and journals: NeurIPS (twice), ICLR, ICML, NeuroImage, Imaging Neuroscience, the Journal of Machine Learning Research (JMLR), and the Learning from Time Series for Health workshop at ICLR.
🗺️ Journey
A research scientist, in motion
Three countries, one through-line: building tools to decode signals from the brain. It started at a 2012 high-school science fair in Mato Grosso do Sul (I was sixteen), went through a cotutelle PhD between Paris-Saclay and UFABC, and continues today at Yneuro, with an honorary affiliation at UC San Diego INC.
PhD in Computer Science (cotutelle)
Cotutelle PhD: Learning Representations of Electroencephalogram using Deep Learning (Paris-Saclay) / Aprendizado de Representações para Eletroencefalografia usando Aprendizado Profundo (UFABC). Advisors: Sylvain Chevallier, Marie-Constance Corsi, Raphael Y. de Camargo. Sandwich period at King's College London with Walter H. L. Pinaya. Funded by INRIA (FR) and CAPES (BR).
-
2026 PhD defenseCotutelle thesis defended February 2026 at Paris-Saclay & UFABC.
-
2024 Geometric Neural Network (JNE)Phase-space SPDNet for BCI-EEG decoding, in the Journal of Neural Engineering with Carrara, Corsi, Papadopoulo.
-
2024 MOABB benchmark studyLargest EEG-based BCI reproducibility study for open science. With Chevallier, Carrara, Guetschel, et al.
-
2024 Euclidean alignment (JNE)Systematic evaluation of Euclidean alignment with deep learning for EEG decoding. Junqueira, Aristimunha, Chevallier, de Camargo.
-
2024 Alljoined dataset (CVPR-W)EEG-to-Image decoding dataset, CVPR 2024 Workshop on Data Curation in Medical Imaging.
-
2024 MOABB Zenodo releaseMother of all BCI Benchmarks: software registry at INRIA, DOI 10.5281/zenodo.
-
2023 Synthetic Sleep EEG (NeurIPS DGM4H)Latent diffusion models for EEG generation, NeurIPS 2023 DGM4H Workshop (Spotlight).
-
2023 Sleep-Energy (IEEE Access)Energy optimization for sleep stage scoring. With Bayerlein, Cardoso, Pinaya, de Camargo.
-
2023 IVA for Motor Imagery (ICASSP)Independent Vector Analysis on EEG-Based Motor Imagery Classification, ICASSP 2023.
-
2023 Braindecode registeredSoftware registration with INRIA, V1.0 (01/08/2023).
-
2023 Braindecode Code-SprintOrganized the European 2023 sprint.
-
2023 King's College London (sandwich)Visiting period under Walter H. L. Pinaya.
-
2023 Started Paris-Saclay legCotutelle PhD enrollment at Paris-Saclay (in addition to UFABC). INRIA scholarship.
-
2022 Glasgow / FGV internData Scientist intern at the University of Glasgow & Fundação Getúlio Vargas.
-
2021 FGV consultantData Science consultant on an IDB-funded urban-data project (Waze car-accident detection in São Paulo). Stack: AWS, SQL, Python, Dash.
-
2020 PhD start (UFABC)Began PhD in Computer Science at UFABC under Raphael Y. de Camargo. CAPES scholarship.
🔬 Featured Manuscript
EEG representation learning
Learning aligned EEG representations with subject-specific encoders
Subject-specific encoders can internalise part of the alignment role usually handled by Euclidean Alignment. Cross-subject decoding performance holds, and head selection becomes the main remaining bottleneck.
🧭 Research Overview
-
P1From EEG Cleaning to DecodingEUSIPCO 2026 · submitted
-
Data resources and platforms
-
P10Alljoined EEG-to-Image DatasetCVPR Workshop 2024
-
Data generation
-
P12Synthetic Sleep EEG with Latent DiffusionNeurIPS DGM4H 2023
-
-
Cross-session and cross-subject alignment
-
P33Subject-Specific Aligned EEG EncodersManuscript 2026 · submitted
-
P9Euclidean Alignment + AugmentationEUSIPCO 2024
-
P11Systematic Euclidean Alignment EvaluationJ. Neural Engineering 2024
-
P14Independent Vector Analysis for MIICASSP 2023
-
-
Online adaptation
-
Decoding architectures
-
P7Geometric Neural Network on Phase SpaceJ. Neural Engineering 2025
-
P15CONCERTO Connectivity Graph NetworkJournée CORTICO 2023
-
P16Holographic EEG Multi-ViewJournée CORTICO 2023
-
P17Sleep-EnergyIEEE Access 2023
-
P25EEG Conformer for P300Preprint (HAL) 2024
-
P26Augmented SPDNet for Motor ImagerySoph.I.A Summit 2023
-
-
Foundation and transfer models
-
P34Masking Geometry for EEG Foundation ModelsManuscript 2026 · submitted to nice conference :)
-
P24Channel Adaptation for EEG Foundation ModelsPreprint (arXiv) 2026 · under review
-
P2OpenEEG-BenchEUSIPCO 2026
-
P3EEG Foundation ChallengeNeurIPS 2025
-
P6General-Purpose Brain Foundation ModelsNeurIPS TSALM 2024
-
P13Cognitive Task Structure with Transfer LearningNeurIPS AI for Science 2023
-
P19Cognitive Task Structure (journal)In re-submission queue 2025 · under review
-
-
P18Largest MOABB Reproducibility StudyPreprint 2025 · under review
-
P4Interpretable MEG Differential DiagnosisHeliyon 2026
-
P21BraindecodeZenodo 2023
- Journal
- Conference
- Workshop
- Abstract
- Software
- Report
📝 Publications (Full List)
2026
EEGDash: An open-source platform for machine learning on public neurophysiological data
Learning aligned EEG representations with subject-specific encoders
What masking geometry works best for EEG foundation models? A controlled evaluation across MAE and JEPA
From EEG Cleaning to Decoding: The Role of Artifact Rejection in MI-based BCIs
Real-Time Continuous EEG Authentication: Streaming Neural Biometrics
Learning Representations of Electroencephalogram using Deep Learning
Toward OpenEEG-Bench: A live community-driven benchmark for EEG foundation models
SPD Learn: A geometric deep learning Python library for neural decoding through trivialization
2025
EEG Foundation Challenge: From Cross-Task to Cross-Subject EEG Decoding
Geometric Neural Network based on Phase Space for BCI decoding
The largest EEG-based BCI reproducibility study for open science: the MOABB benchmark
Uncovering and improving the structure of cognitive tasks with transfer learning
2024
General-Purpose Brain Foundation Models for Time-Series Neuroimaging Data
What is the best model for decoding neurophysiological signals? Depends on how you evaluate
Combining Euclidean Alignment and Data Augmentation for BCI decoding
A systematic evaluation of Euclidean alignment with deep learning for EEG decoding
Evaluation of the Electroencephalogram Conformer for the P300 Signal
2023
Synthetic Sleep EEG Signal Generation using Latent Diffusion Models
Evaluating the structure of cognitive tasks with transfer learning
Applying independent vector analysis on EEG-based motor imagery classification
Sleep-Energy: An Energy Optimization Method to Sleep Stage Scoring
Augmented SPDNet: Second-Order Neural Network for Motor Imagery-Based BCI
Distribuição de Newcomb-Benford, Mudanças de Bases e Aplicações Eleitorais
2020
2019
📖 Education
-
09/2020 – 02/2026 PhD in Computer Science
Cotutelle between Université Paris-Saclay 🇫🇷 and UFABC 🇧🇷. Advised by Sylvain Chevallier, Marie-Constance Corsi, and Raphael Y. de Camargo. Thesis: Learning Representations of Electroencephalogram using Deep Learning.
-
2016 – 2020 Double BSc in Computer Science & Science and Technology
Center for Mathematics, Computing, and Cognition, Federal University of ABC (UFABC), Brazil 🇧🇷.
💻 Work Experience
-
2026 → Research Scientist, Yneuro
France 🇫🇷. Tools for EEG decoding and foundation models on neural signals.
-
2026 → Honorary Research Associate, UC San Diego (INC)
Institute for Neural Computation, USA 🇺🇸.
-
03/2022 – 06/2022 Data Scientist Intern, University of Glasgow / FGV
United Kingdom 🇬🇧.
-
03/2021 – 08/2021 Data Scientist Intern, Getúlio Vargas Foundation (FGV)
Brazil 🇧🇷.
-
07/2014 – 12/2015 Research Intern (Computer Vision), Dom Bosco Catholic University
Brazil 🇧🇷. INOVISÃO lab during high school. I published two papers :)
👥 Mentorship
Students I was lucky to work with and mentor:
-
Léo Burgund
Master student → YneuroUniversité Paris-Saclay (M2 Mathematics & AI), now Machine Learning Researcher at Yneuro.
-
Tom Mariani
Master student → YneuroMVA, ENS Paris-Saclay / Mines, now Research Scientist at Yneuro.
-
Aman Jaiswal
Master studentUC San Diego, MS in Computer Science.
-
Kuntal Kokate
Master studentUC San Diego, MS in Machine Learning & Data Science (ECE).
-
Jose Mauricio
Master studentFederal University of ABC, Computer Science.
-
Taha Habib
Undergrad → MasterUniversité Paris-Saclay, now a master student.
-
Gustavo H. Rodrigues
Undergrad → MasterUniversidade de São Paulo (USP), now a master student at USP.
-
Bruna Junqueira
Undergrad → MasterUSP, now in the Mathématiques, Vision, Apprentissage master at Université Paris-Saclay.
-
Alexandre Janoni
Undergrad → IndustryFederal University of ABC, now at Hospital Albert Einstein.




























