Duminda Ranasinghe
Senior Scientist II at Montai Health
Cambridge, Massachusetts, United States
Summary
👤
SeniorDuminda Ranasinghe is an Associate Director in Cambridge, MA with nine years of experience applying machine learning and computational chemistry to accelerate early drug discovery and lead optimization. He leads a cross-functional computational biology and ML team at Montai Therapeutics, building ML-driven ADMET predictors, response-surface models, and CADD pipelines that have directly informed HTS and lead nomination. His background marrying quantum chemistry and software is unique: he contributed back-end enhancements to the widely used RMG-Py project, adding support for Orca and advanced post-processing of high-level quantum methods. Previously he developed kinetic and solvation ML models at MIT and produced automated reaction mechanism tools for exascale simulations, demonstrating strengths at the intersection of theory, computation, and scalable engineering. Collected experience with CI/CD, cloud-accelerated training on AWS, and interactive dashboards ensures his work translates into operational drug discovery impact.
9 years of coding experience
Github Skills (6)
quantum-chemistry10
computational-chemistry10
python10
p29
mpeg9
thermodynamics9
Programming languages (2)
Github contributions (5)
ReactionMechanismGenerator/RMG-Py
Apr 2019 - Dec 2020
Python version of the amazing Reaction Mechanism Generator (RMG).
Role in this project:
Back-end Developer / Computational Chemist
Back-end Developer / Computational ChemistContributions:3 reviews, 29 commits, 5 PRs in 1 year 7 months
Contributions summary:Duminda primarily contributed to the Arkane module, focusing on implementing and expanding the functionality to parse and analyze output files from quantum chemistry software. They added support for Orca, parsing its output, including energies and geometries. Further contributions included the parsing of output files for MP2, double hybrid, CCSD, and CCSD(T) methods within Gaussian calculations, enhancing the capabilities of the tool for thermodynamic analysis. These additions indicate a focus on expanding the types of quantum chemical data that can be used within the overall project.
mechanismchemical-engineeringchemistrypythonreaction
fl65inc/chemprop
Mar 2021 - Dec 2022
Message Passing Neural Networks for Molecule Property Prediction
Contributions:4 PRs, 2 pushes, 4 branches in 1 year 8 months
property-predictionpredictiondeep-learningneural-message-passingmolecule
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.