Skip to content
View mondalsou's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report mondalsou

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
mondalsou/README.md

Hi, I'm Sourav Mondal, PhD πŸ‘‹

AI/ML Scientist | Computational Chemist | Drug Discovery

Portfolio LinkedIn GitHub Email


🧬 About Me

Computational scientist with 10+ years of experience in cross-disciplinary research spanning Drug Discovery and Materials Modeling. Currently at Aganitha Cognitive Solutions, leveraging quantum mechanical calculations and deep learning to accelerate small molecule research.

  • πŸ”¬ Building end-to-end pipelines for crystal structure prediction, conformer generation, pKa estimation, and ADMET property calculations
  • πŸ“š Published in Nature Computational Materials, JACS, and Nano Letters
  • πŸŽ“ PhD from JNCASR Bangalore | Postdoc at Trinity College Dublin

πŸ”­ Current Work at Aganitha Cognitive Solutions

Project Description
Crystal Structure Prediction Graph Neural Networks for rapid polymorph screening of drug-like molecules
Conformer Generation AI-driven platform enabling environment-aware conformer generation for advanced drug discovery applications
pKa Estimation Two complementary modeling approachesβ€”QM-based and GNN-basedβ€”for pKa estimation under non-aqueous conditions
ADMET Prediction Physics-based QM platform for high-accuracy prediction of key thermodynamic properties: solubility, logP, and solvation energy

πŸš€ Featured Projects

Project Stack Highlights
Drug Discovery Triage React Β· FastAPI Β· RDKit Real-time ADMET predictions, QED scoring, PAINS alerts, live demo
Solubility Dual-Graph GNN PyTorch Β· MPNN Β· Cross-Attention RΒ²=0.90 on 100K+ solute-solvent pairs (BigSolDB 2.0)
Lead Optimization Agent Claude AI Β· RDKit Β· Streamlit AI agent loop for iterative structural optimization with property tracking
GNN Bind Optimizer PyTorch Β· PyG Β· REINFORCE Β· MLflow Heterogeneous GNN for binding affinity + RL-based pocket-conditioned molecule generation
Formulation Bayesian Optimization BoTorch Β· GPyTorch Β· RDKit Β· Streamlit Bayesian optimization pipeline for pharmaceutical formulation design β€” multi-objective GP surrogate with acquisition function tuning

πŸ› οΈ Technical Skills

AI/ML & Deep Learning

Gen AI PyTorch TensorFlow GNN Transformers AI Agents Scikit-learn Reinforcement Learning Transfer Learning Multi-task Learning LSTM

Computational Chemistry

CP2K ORCA psi-4 PySCF DFT/SAPT ML Force Fields

Cheminformatics

RDKit ASE SMILES ADMET Docking CADD Binding Affinity Lead Optimization

Programming & Tools

Python Fortran Docker Azure Git n8n Streamlit SQL Server MLflow


πŸ“š Featured Publications

Year Publication Journal
2025 The spin phonon relaxation of single molecules magnet in the presence of strong exchange coupling ACS Cent. Sci.
2023 Spin-phonon decoherence in solid-state paramagnetic defects from first principles npj Comput. Mat.
2022 Unravelling the contributions to spin-lattice relaxation in Kramers single-molecule magnets J. Am. Chem. Soc.
2022 Identification and Manipulation of Defects in Black Phosphorus J. Phys. Chem. Lett.
2020 Direct Observation of the Reduction of a Molecule on a Nitrogen pair in Doped Graphene Nano Lett.

πŸ“Š GitHub Stats

GitHub Stats

Top Languages


πŸŽ“ Education

  • PhD in Computational Material Science - JNCASR, Bangalore (2015-2020)
  • M.Sc in Chemistry - IIT Guwahati (2012-2014)

πŸ“« Let's Connect

πŸ“ Hyderabad, India | πŸ“§ souravchembwn@gmail.com | πŸ“± +91 9740851654

Portfolio CV


Computational Scientist | AI/ML Engineer | Building intelligent systems at the intersection of science and technology

Pinned Loading

  1. drug-discovery-triage drug-discovery-triage Public

    πŸ”¬ Full-stack drug discovery triage app | React + FastAPI + RDKit | ADMET predictions, QED scoring, PAINS alerts | Live demo

    TypeScript

  2. solubility-dual-graph-gnn solubility-dual-graph-gnn Public

    Dual-graph interaction GNN for predicting molecular solubility (logS) from solute-solvent SMILES using cross-attention and BigSolDB 2.0

    Jupyter Notebook

  3. lead-optimization-agent lead-optimization-agent Public

    AI agent (Claude + RDKit + Streamlit) for iterative lead optimization β€” proposes structural changes, scores ADMET/QED/BBB properties, and tracks the candidate journey visually

    Python

  4. gnn-bind-optimizer gnn-bind-optimizer Public

    Heterogeneous GNN for protein-ligand binding affinity prediction + REINFORCE molecular generator Β· EGFR kinase Β· SQL Server Β· MLflow Β· Streamlit

    Jupyter Notebook

  5. quantum-pharma-lab quantum-pharma-lab Public

    Quantum pharma demo using QAOA and Qiskit to select drug candidates from molecular datasets, with classical ML baselines and an interactive wave-interference explainer.

    Jupyter Notebook

  6. formulation-bayesian-optimization formulation-bayesian-optimization Public

    Model-based DoE and Bayesian optimization for pharmaceutical formulation: GP-guided Q45 optimization with EI/LogEI, plus qNEHVI multi-objective trade-offs (Q45, hardness, friability).

    Jupyter Notebook