Ramaravind M
Residence Advisor, University Family Housing at University of Toronto
Old Toronto, Ontario, Canada
Summary
🤩
Rockstar🎓
Top SchoolRamaravind M is a PhD student at the University of Toronto and researcher focused on the safe, responsible deployment of AI at the intersection of HCI, NLP, and FAccT, with eight years of industry and research experience. He builds practical tools and theoretical frameworks to surface the tensions and technical assumptions behind AI systems, and his current work probes the reasoning-like behaviors of LLMs that shape user trust and decision-making. An active contributor to open-source ML explainability—having implemented and refined PyTorch-based counterfactuals in the notable DiCE library—he pairs rigorous research with hands-on engineering. His background spans Microsoft Research India, industry data science roles, and advanced AI study at KU Leuven, reflecting a blend of applied engineering, academic depth, and a focus on fairness in socio-technical systems.
8 years of coding experience
1 year of employment as a software developer
Advanced Masters Artificial Intelligence, Advanced Masters Artificial Intelligence at KU LeuvenBachelor's Degree Instrumentation and Control, Bachelor's Degree Instrumentation and Control at National Institute of Technology, TiruchirappalliDoctor of Philosophy - PhD Information Science/Studies, Doctor of Philosophy - PhD Information Science/Studies at University of Toronto
English, Saurashtra, Tamil, Hindi
Github Skills (4)
machine-learning10
python10
pytorch10
deep-learning9
Programming languages (5)
Github contributions (5)
interpretml/DiCE
Jun 2019 - Feb 2021
Generate Diverse Counterfactual Explanations for any machine learning model.
Role in this project:
ML Engineer
ML EngineerContributions:3 reviews, 160 commits, 25 PRs in 1 year 8 months
Contributions summary:Ramaravind primarily contributed to the development of the DiCE library, focusing on the implementation and refinement of the PyTorch-based counterfactual explanation generation functionality. Their work involved implementing and refactoring code within the `dice_pytorch.py` file. They also made code adjustments and ensured that the library could correctly handle a two-output-node binary classifier. Furthermore, the user fixed some issues and implemented posthoc sparsity enhancement in the code.
explainable-mlcounterfactual-explanationsdiversedataminingxai
raam93/Taxi-Trip-Analysis
Dec 2017 - Nov 2020
Contributions:2 PRs, 5 pushes, 3 branches in 2 years 11 months
data-analyticsanalyticstotaltripsairport
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.