Hiring for Data Scientist? Start here.
IIT Madras Data Science & AI, GPA 7.8+. Statistical modelling, honest evaluation, and experiments designed to survive contact with production.
Top 25
Riot x AWS hackathon, globally
7.8+
GPA, IIT Madras
25+
classifiers built and evaluated
About me
I am Vansh Bordia, a data scientist at IIT Madras with a GPA above 7.8. My interest is the unglamorous part of modelling: what the baseline is, whether the split leaks, and whether the number survives a new season of data.
I build models for people who use them daily, not for leaderboards. Playstyle classifiers for coaching staffs, diffusion work I read early and built on, and a bias toward shipping the smallest honest version.
Why me
I care less about a score inside a notebook and more about whether a model still holds three months later: real baselines, leak-free splits, evaluation I would defend in a review.
Data Science & AI at IIT Madras, taught by building: domain-specific classifiers on real esports telemetry, experiments with baselines that matter, and evaluation setups that catch leakage before it flatters me. I care more about whether a model holds up outside the notebook than how it scores inside one.
Top 25 globally, Riot x AWS hackathon
25+ playstyle classifiers built on esports telemetry
BSc Data Science & AI, IIT Madras, GPA above 7.8
Role stack
Python / PyTorch / Statistics / Experimentation / A/B Testing / Scikit-Learn / Pandas / Feature Engineering / Model Evaluation / Causal Inference / LLM Fine-Tuning / RAG Pipelines / Data Visualisation
Relevant work
Esports intelligence platform, acquired by a pro org
Type-safe Python library for Valorant data, published on PyPI
Interactive SQL schema visualizer, 100% client-side
Contact
Need models that survive production? Talk to me.