Francisco H
Science Research Engineer at Google DeepMind
Stony Stratford, England, United Kingdom
Summary
👤
Senior🎓
Top SchoolFrancisco H is a Science Research Engineer at Google DeepMind with nine years of experience at the intersection of AI and quantitative biology. He combines a strong academic background (PhD in Zoology/Biophysics from Cambridge and dual MAs in Mathematics and Physics) with hands-on ML and computer vision engineering developed during postdoctoral work at Champalimaud Foundation. His work spans self-supervised learning, explainable deep models of collective animal behaviour, and production-grade tracking software used by multiple labs to monitor hundreds of animals. An active open-source contributor, he improved statistical evaluation in the popular mlxtend library by implementing robust permutation tests and fixing subtle refactor bugs. He also has a track record of making complex tools accessible—publishing packages and notebooks to let non-specialists analyse trajectories—and teaching ML and CV to doctoral students. Based in Stony Stratford, he blends rigorous scientific thinking with practical engineering to turn biological questions into reproducible AI systems.
9 years of coding experience
6 years of employment as a software developer
MPhil, Biological Science, MPhil, Biological Science at University of CambridgeLicenciado en Matemáticas (M.A. Mathematics), Licenciado en Matemáticas (M.A. Mathematics) at Universidad de Valladolid
Spanish, French, Portuguese, German, English
Github Skills (10)
machine-learning10
python10
data-science10
numpy10
scikit-learn9
scikit9
unsupervised-learning8
supervised-learning8
testing7
data-mining6
Programming languages (10)
Github contributions (5)
rasbt/mlxtend
Jan 2021 - Jan 2021
A library of extension and helper modules for Python's data analysis and machine learning libraries.
Role in this project:
Data Scientist
Data ScientistContributions:5 commits, 1 PR, 1 comment in 1 day
Contributions summary:Francisco contributed to the `mlxtend` library by implementing and refactoring permutation tests. They added functionality for paired permutation tests and incorporated changes to existing tests. Furthermore, the user addressed style issues and fixed bugs introduced during the refactoring process. Their work focused on improving the functionality and reliability of the statistical evaluation module within the library.
supervised-learningpythondata-analysisdata-scienceunsupervised-learning
fjhheras/trajectorytools
Nov 2017 - Nov 2021
A library with some utils to study and plot 2D trajectories developed by members of "de Polavieja lab" (Champalimaud Research, Lisbon, Portugal). Used in the study of animal collective behaviour.
Contributions:8 releases, 16 reviews, 271 commits in 4 years
plotmembersutilsanimallisbon
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.