CV
General Information
| Full Name | Roshan Balaji |
| Languages | English, Tamil, Malayalam, Hindi |
Education
- 2021 - Present
PhD
Indian Institute of Technology, Madras
- Prime Minister's Research Fellow
- Wadhwani School of Data Science and Department of Biotechnology
- Teaching Assistant:
- Big Data and Biological Networks, Multivariate Data Analysis, Industrial AI Laboratory, Programming for Biological Engineers, Bio-process Control, Current Topics in Synthetic Biology
- 2017 - 2021
Bachelor of Technology
National Institute of Technology, Calicut
- Gold Medalist (CGPA 9.08)
- Thesis: Cancer subtype prediction using histopathology images
- Core Courses:
- Bioinformatics, Bio-signal Processing, Biostatistics, Bioreactor Design and Analysis
Experience
- Nov 2025 - Present
Open Source Contributor
OpenFold Initiative
- May 2025 - Nov 2025
Research Intern
Mila - Quebec AI Institute, Montreal, Canada
- Goal-aware retrosynthesis planning, with Prof. Sarath Chandar
- June 2022 - July 2022
Teaching Assistant
ACM - RBCDSAI Summer School
- Lab sessions on mathematical foundations of data science
- June 2020 - Nov 2020
Research Intern
Boltzmann Labs, Bangalore, India
- Protein-ligand binding site prediction and de novo molecule generation using deep learning
- Implementing and testing algorithms on large datasets for drug discovery using the PyTorch Lightning framework
- April 2019 - July 2019
Research Intern
Swagene, Chennai, India
- Research and Development into novel molecular diagnostics focusing on cancer
- Executed research projects critically and collaborated with labs while maintaining precise documentation
Publications
- A. A. P. M. Shums*, R. Balaji*, N. P. Bhatt, "Leak-CURBER - A Leakage-Controlled Multimodal Evaluation Benchmark for Enzymatic Reaction Tasks," NeurIPS 2026, Evaluations and Datasets Track. (*equal contribution)
- R. Balaji, J. Bobby, N. P. Bhatt, "Functional Groups are All you Need for Chemically Interpretable Molecular Property Prediction," Journal of Chemical Information and Modeling, 2026.
- A. Sinha, R. Balaji, N. P. Bhatt, "BiCA - Effective Biomedical Dense Retrieval with Citation-Aware Hard Negatives," AAAI 2026, Oral (top 5%).
- K. Chitsaz*, R. Balaji*, Q. Fournier, N. P. Bhatt, S. Chandar, "NovoMolGen - Rethinking Molecular Language Model Pretraining," ICML 2025 GenBio Workshop. (*equal contribution)
- See the publications page for the full list.
Open Source Projects
- 2020-now
MolPro
- Python package to accelerate drug discovery using deep learning algorithms
Technical Skills
-
Programming Languages:
- C++, Python, R, awk, MATLAB
-
Libraries and Frameworks:
- PyTorch, pytorch-geometric, Keras, scikit-learn, NLTK, networkx, Django, REST
-
Miscellaneous:
- Linux OS, Git
Positions of Responsibility
- Organizer and Program Chair: GCLR Workshop at AAAI 2023 and 2026
- Reviewer: International Conference on Intelligent Systems for Molecular Biology (ISMB)
- Research Legislator: Department of Biotechnology, IIT Madras
- Mentor: Teach in 10, IIT Madras
- Member: RBCDSAI and IBSE, IIT Madras
- Technical Affairs Secretary: School of Biotechnology, NIT Calicut (Aug 2018 - Aug 2019)