Pere-lluís Cabot
Postdoctoral Researcher at Meta
Paris, Ile-de-France
Summary
👤
Senior🎓
Top SchoolPere-lluís Cabot is a multilingual NLP researcher and software engineer with eight years of experience, currently a postdoctoral researcher at Meta FAIR specializing in multilingual representations and modeling. He completed a PhD in Artificial Intelligence at Sapienza after a cum laude MSc in AI from the University of Amsterdam and a double degree in Physics and Mathematics, blending strong theoretical foundations with practical ML engineering. His work spans academic publications (including EMNLP) and industry projects, from building QA systems with BERT/XLNet to contributing ML evaluation metrics and PyTorch examples to the widely used LIT interpretability tool. He has experience deploying transformer-based multi-task models for tasks like populist rhetoric analysis and has a track record of making research tools more robust and framework-agnostic. Based in Paris, he combines rigorous research, hands-on code contributions, and interdisciplinary collaborations across psychology and EU-funded projects.
8 years of coding experience
6 years of employment as a software developer
Erasmus Exchange, Nuclear Physics and Financial Stochastics, Erasmus Exchange, Nuclear Physics and Financial Stochastics at Humboldt-Universität zu BerlinDoctor of Philosophy - PhD, Artificial Intelligence, Doctor of Philosophy - PhD, Artificial Intelligence at Sapienza Università di RomaMaster's degree, Artificial Intelligence, 8.7 GPA (Cum Laude), Master's degree, Artificial Intelligence, 8.7 GPA (Cum Laude) at Universiteit van AmsterdamDoble grado en Física y Matemáticas (Double degree in Physics and Mathematics), Doble grado en Física y Matemáticas (Double degree in Physics and Mathematics) at Universitat de Barcelona
English, German, Catalan, Spanish, griego, Italian
Stackoverflow
419reputation
6kreached
3answers
6questions
Github Skills (14)
pytorch10
machine-learning10
python10
natural-language-processing10
visualizations9
scikit-learn9
scikit9
visualization9
transformers9
transformer9
tensorboard6
hyperledger6
hyperledger-fabric6
hyperledger-composer6
Programming languages (8)
Github contributions (5)
PAIR-code/lit
Sep 2020 - Sep 2020
The Learning Interpretability Tool: Interactively analyze ML models to understand their behavior in an extensible and framework agnostic interface.
Role in this project:
ML Engineer
ML EngineerContributions:1 review, 5 commits, 4 PRs in 8 days
Contributions summary:Pere-lluís primarily contributed to the development and modification of machine learning model evaluation metrics within the LIT framework. They fixed a bug related to edge cases when evaluating regression metrics with limited data samples, ensuring the system's robustness. Additionally, the user added a PyTorch example to the repository, demonstrating the integration of a custom PyTorch model with the LIT interface, expanding the framework's support for different machine learning backends and providing examples for users. Furthermore, style guide adaptations and import removal was addressed.
ml-modelspythoninterpretabilitydata-sciencedeep-learning
Babelscape/rebel
Oct 2021 - Nov 2022
REBEL is a seq2seq model that simplifies Relation Extraction (EMNLP 2021).
Contributions:1 review, 23 commits, 2 PRs in 1 year
nlpextractionnatural-language-generationseq2seqseq2seq-model
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.