Final-year Data Science & Applications student at IIT Madras. I take models out of notebooks — into pipelines, papers, and products.
I'm a final-year student in the Data Science and Applications programme at IIT Madras, working across ML/DL, MLOps, and backend engineering. My research has taken me from graph neural networks for drug discovery to anomaly detection in smart grids — and into a co-authored publication along the way.
Alongside coursework, I mentor as a teaching assistant to over 5,000 students — mostly a good reminder that explaining an idea clearly is harder than understanding it.
A short summary of a few projects that made it from idea to something real.
Applied graph transformers to compute drug similarity matrices across 267 drugs during a summer internship — work later co-authored into a published paper.
Built centralised ML pipelines to detect anomalies in smart grid data, benchmarking logistic regression against CNN, RNN, and ANN architectures.
Trained ensemble models on 45K+ marketing records with severe class imbalance, reaching an F1 score of 0.9477 via SMOTE and hyperparameter tuning.
Endpoint-wise pytest suites for auth, RBAC, and simulation APIs. Code reviewer on 30+ merged PRs.
Full-stack role-based parking system in VueJS with Redis caching and scheduled jobs.
Multi-user request management app in Flask with notifications and token-based access.
Ran tutorials, set and graded assessments, and curated coursework for 1,200+ students.
Built centralised ML pipelines for smart grid anomaly detection.
Graph transformer-based drug similarity analysis — later a co-authored publication.
Mentored 5,000 students across three cohorts of Mathematics for Data Science–I.