Tom Sercu
Vice President, AI And Engineering at Biohub
New York, New York, United States
Summary
🤩
Rockstar🎓
Top SchoolTom Sercu is a seasoned AI and engineering leader with 12 years of experience building research-driven teams at the intersection of deep learning and biology. Currently Vice President, AI and Engineering at Biohub, he previously led Meta FAIR’s Protein team where he shipped landmark models like ESM, ESMFold and ESM Atlas and grew the group to ~15 engineers and scientists. He co-founded EvolutionaryScale to translate protein language-model advances into product, and his open-source contributions include improving usability and adding pretrained models and contact prediction to the widely used facebookresearch/esm repo. Tom’s background spans top research labs (Meta, IBM, NYU) and rigorous physics and data science training from Ghent University and NYU, giving him a strong foundation in both theory and applied ML. Colleagues describe him as a pragmatic researcher-engineer who moves models from publication to production while keeping developer experience and documentation front and center.
12 years of coding experience
10 years of employment as a software developer
Master of Science (MS) Data Science, Master of Science (MS) Data Science at New York UniversityMaster of Science (M.Sc.) Engineering Physics/Applied Physics, Master of Science (M.Sc.) Engineering Physics/Applied Physics at Ghent UniversityErasmus exchange Physics, Erasmus exchange Physics at University of Stuttgart
Dutch, English, German, French
Stackoverflow
13reputation
117reached
0answers
1question
Github Skills (9)
bioinformatics10
transformer-models10
pytorch10
machine-learning10
deep-learning10
python10
documentation9
plot6
matplotlib6
Programming languages (7)
Github contributions (5)
facebookresearch/esm
Sep 2020 - Jan 2023
Evolutionary Scale Modeling (esm): Pretrained language models for proteins
Role in this project:
ML Engineer
ML EngineerContributions:51 reviews, 88 commits, 56 PRs in 2 years 4 months
Contributions summary:Tom made significant contributions to the project by modifying existing code to improve the usability of the `extract.py` script. Further work was done to refine the README and notebook documentation, including adding example code for variant prediction. The user also added new pre-trained models and contact prediction functionality. These changes suggest a focus on model usability, documentation, and extending the capabilities of the ESM models.
scale-modelingproteinspretrained-language-modelsesmscale
tomsercu/Xtrac
Oct 2013 - May 2014
Contributions:101 commits in 6 months
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.