top of page
f1bd1ac1-93ff-44b8-afc0-998df84efcd7.JPG

Shreyans Gupta

Junior at Duke University studying Data Science
Based in Durham, North Carolina

Screen Shot 2024-07-23 at 11.24.48 AM.png

About Me

Graduating in May 2026 with a Double Major in Math & Statistics (Concentration in Data Science). 4+ years of experience with predictive modeling, machine learning, and natural language processing. I enjoy taking data and extracting actionable insights for stakeholders and businesses, driving product innovation and strategic decision-making.

In my free time, I powerlift, meal prep, watch the NBA, and solving puzzles!

Skills

Currently Learning

  • GitHub
  • LinkedIn

grew my edtech start-up to 25k students                   published a math book                   i can solve a rubik's cube < 15 seconds                   i represented India in a Math Olympiad

(ask me to show you)

Projects 

astrachanSearch.png

Google Search Autocomplete and Visualizer

The AutoComplete project utilizes weighted prefixes and binary search to suggest relevant completions based on the user's search input. It efficiently processes user queries to provide accurate and fast autocomplete suggestions, enhancing user experience in text-based searches. It also updates the visualization in real-time with each letter.

Read more 

imageforboard.png

The Board-Game-Popularity project conducts a comprehensive analysis to predict the popularity of board games based on factors such as ratings, complexity, and playtime. Using machine learning techniques like dimensionality reduction, feature engineering, and multi-regression models, it determines that difficult, fantasy-themed miniature games are statistically the most popular.

Read more 

durhamGraph_edited_edited.png

Shortest Path Finder on U.S Map

The project implements a graph processor and leverages Dijkstra's algorithm and k-NN search to calculate and visualize the shortest route between two cities. It asks the user for a city of departure and a city of arrival and calculates the distance, the processing time, and also shows the route on a U.S Map.

Read more 

image.png

This project aims to demystify the system of deciding the NBA's clutch player of the year (player who performs well when it matters). It employs methodologies such as feature weights and regression analysis to create a holistic metric for evaluating clutch performance. Uses Shiny R to display the top clutch performers, allowing a user to search by year, specific box score preference, and minimum minutes played.

Read more 

What I'm working on right now: Building the NBA Schedule That Saves the Most Amount of Fuel

Contact Me!

bottom of page