Arian Jamasb
Principal Machine Learning Scientist at Genentech
Basel, Basel-City, Switzerland
Summary
🤩
Rockstar🎓
Top SchoolArian Jamasb is a Principal Machine Learning Scientist with nine years of experience applying deep learning and graph-based ML to structural biology and drug discovery. He holds a PhD from the University of Cambridge and has held research roles across industry and academia including Genentech, Google X, MILA, and Relation Therapeutics. Arian’s work spans production ML at Genentech and open-source contributions—most notably to BioPandas and Graphein—where he implemented mmCIF parsing and protein graph construction utilities that bridge molecular structures and dataframes/graph models. He blends hands-on backend development with research rigor, having worked on synaptic-resolution connectomics and graph ML for drug development. Based in Basel, he is comfortable moving projects from novel research to deployable systems and often focuses on the less obvious engineering challenges such as data parsers and feature plumbing that make advanced models usable.
9 years of coding experience
5 years of employment as a software developer
Doctor of Philosophy - PhD Artificial Intelligence Structural Biology & Drug Discovery, Doctor of Philosophy - PhD Artificial Intelligence Structural Biology & Drug Discovery at University of CambridgeBachelor of Science (B.Sc.) Biochemistry, Bachelor of Science (B.Sc.) Biochemistry at Imperial College LondonThe Perse School, Cambridge
English, Norwegian, Swedish, French
Github Skills (5)
bioinformatics10
data-structures10
networkx10
python10
data-structure10
Programming languages (13)
Github contributions (5)
a-r-j/graphein
Aug 2019 - Jan 2023
Protein Graph Library
Role in this project:
Back-end Developer & Data Scientist
Back-end Developer & Data ScientistContributions:19 releases, 116 reviews, 877 commits in 3 years 5 months
Contributions summary:Arian's commits involve the initial implementation of a class for protein graph construction, including methods for initializing the graph with nodes, and features. They also demonstrate the implementation of methods for incorporating different edge types such as peptide bonds and hydrogen bonds. Additionally, the user's work indicates a focus on adding data from sequence analyses, suggesting a data science and bioinformatics focus. This demonstrates work in graph-based machine learning applied to protein structure data.
graphproteinprotein-structuredeep-learningpytorch
BioPandas/biopandas
Mar 2022 - Jan 2023
Working with molecular structures in pandas DataFrames
Role in this project:Back-end Developer
Back-end DeveloperContributions:5 releases, 36 reviews, 87 commits in 10 months
Contributions summary:Arian appears to be focused on the development of the mmcif parser, with initial work on parsing mmcif files. Their commits involve creating a parser class, refactoring existing code, and implementing util methods for use within the project. The user also worked on the core methods for handling carbon atoms.
pandas-dataframemolecular-structurespdbmol2protein-structure
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.