Hello!

I'm Tanmay Kapure



A Gradute Student in Data Science at New Jersey Institute of Technology

My works

Name: Tanmay Kapure

Job Role: Data Science Student

Experience: 1 year

Address: Newark, NJ

Skills

LLM 70%
Agentic AI 70%
SQL 80%
PYTHON 85%
Data Visualization 80%
Statistical Analysis 85%
Machine Learning 85%
Deep Learning 80%
Natural Language Processing 80%

About

About Me

Welcome! I'm Tanmay Kapure, a dedicated and results-oriented Data Scientist with a passion for uncovering insights and driving innovation through data. Currently pursuing a Master of Science in Data Science at the New Jersey Institute of Technology, where I blend a robust academic foundation with hands-on industry experience. Through high-impact roles— including working on projects for reputed organizations like U.S Army and NASA —I've consistently driven measurable improvements in efficiency and accuracy. My expertise spans advanced statistical modeling, machine learning, and data visualization, complemented by proficiency in cutting-edge tools like AWS, Transformers, and anomaly detection systems. Passionate about solving complex challenges, I thrive on turning large datasets into strategic solutions and leading multidisciplinary projects that push the boundaries of innovation.

  • Profile: Data Science & Analytics
  • Domain: Retail, Ecommerce, Physics & Finance
  • Education: Bachelors in Artificial Intelligence ; Masters in Data Science
  • Language: English, Hindi, Marathi
  • BI Tools: Microsoft Power BI, Looker & Tableau
  • Other Skills: Cloud, PySpark, Excel, Git, JIRA, MySQL & GenAI
  • Interest: Traveling, Fitness, Self Improvement

0 +   Projects completed

LinkedIn

Resume

Resume

Professional Data Scientist driving business strategies through data-driven insights. Proven expertise in data science, statistical analysis, machine learning algorithms and project management.

Experience


Oct 2024 - Present

Modelling and Prediction Analyst (U.S ARMY Funded Research Project)

New Jersey Institute of Technology

Worked with Senior Professors and Researchers of U.S Army to research and utlizie advanced methods to predict fatigue strength of Metals

  • Developed deep learning models to predict metal fatigue strength, incorporating advanced architectures like Transformers.
  • Improved prediction accuracy, reducing Mean Relative Error (MRE) from 140% to 60%.
  • Reviewed and implemented research papers, applying state-of-the-art deep learning techniques to fatigue strength prediction.

1
2
Mar 2024 - Oct 2024

Graduate Research Assistant (NASA and NSF Funded Project)

Center of Solar and Terrestrial Research, NJIT

The CSTR is an international leader in solar and terrestrial physics, with interest in understanding the effects of the Sun on the geospace environment.

  • Enhanced and analyzed 50,000+ raw solar surface images using advanced image processing techniques.
  • Created more than 500 high-quality videos for research and educational purposes, providing valuable insights.
  • Collaborated with a multidisciplinary team of 15+ researchers to facilitate extraction of information from solar imagery.
  • Conducted software testing, debugging, and code integration for running ray tracing simulations parallelly on HPC cluster.

Mar 2023 - Sep 2023

Machine Learning Intern

Everlytics Data Science Pvt Ltd

A profitable startup based in singapore, which provides software solutions and services that help companies

  • Implemented data extraction pipelines and job scheduling piplines using AWS Glue
  • Developed robust anomaly detection model (Isolation Forest) for anomaly detection in production pipeline. Got model performance of 92%.
  • Debugged producer-consumer connection issues data pipelines on Apache Airflow. Transformed ~ 1 million rows of data weekly.
  • Optimized data parsing algorithms to resolve input CSV length issues

3
4
Nov 2022 - Mar 2023

Project Intern

Jubilant Biosys Pvt Ltd

Jubilant Biosys is a Contract Research, Development and Manufacturing Organization in India providing comprehensive drug discovery and research services worldwide.

  • Optimized forecasting models (Random Forest, ARIMA) to 91% accuracy, reducing overstocking by 20% and understocking by 15%.
  • Designed 5 interactive dashboards in Tableau, increasing data visibility by 40% and enabling real-time KPI tracking.
  • Executed A/B testing to optimize inventory strategies, reducing stockouts by 10% and improving order fulfilment rates by 5%.
  • Partnered cross-functionally with Product, Engineering teams to integrate predictive models into the supply chain system.

Education

Education

My academic journey and achievements.

2023-Present

MS in Data Science

New Jersey Institute of Engineering and Technology

GPA 3.91 / 4.0

2019-2023

B-Tech in Artificial Intelligence

Rashtriya Sant Tukdoji Maharaj Nagpur University

GPA 3.74 / 4.0

Download CV

Projects

Projects

Below are my personal projects in Python, ML, Gen AI and Full Stack Development.

RAG based AI Chatbot for Intelligent Document Retrieval

Built an end-to-end RAG system leveraging LangChain, FastAPI, and Pinecone, enabling context-aware responses from LLMs. Enhanced document processing & chunking using sentence transformers and made a Gradio webapp to interact with model. Orchestrated the webapp on Hugging Face Spaces with Docker & CI/CD, ensuring scalability & real-time query handling.

Browser AI Agent

Created browser agent to navigate, extract info from web as per natural language instructions using Langgraph, Gemini, Playwright. Ability to bypass captcha, scrape websites, Human in the Loop feature to seamlessly fulfill browser tasks. Smart LLM assisted navigation to successfully handle dynamically changing web layouts.

Voice AI Interview Assistant

A full stack Voice AI based end to end web application agent to simulate mock interview using VAPI AI and Gemini API's. Intelligent Chatbot that gives feedback based on candidate's performance and helps them evaluate their preparation for Interviews. Responsive and beautiful frontend based on Next.js and Tailwind CSS, while PostgreSQL for database and Firebase for authentication, hosted on Vercel.


Supply chain management using quantum computing

Compared the performance of Quantum Machine Learning algorithms vs the classical ML agorithms for solving the backorders problem and Optimal vehicle route prodection in supply chain management

Exoplanet prediction using NASA's Kepler Dataset

Performed exoplanet exploration using NASA's keper telescope dataset from NASA's exoplanet archive

Emotion recognition using NLP

Predicted various sentimens using BILSTM-RNN with an accuracy of 94 % on emotions data


Natural Language to SQL conversion using LLM

We Used RoBERTa LLM for converting Natural language sentences into SQL queries

Retail Association rule mining

Developed an ML model to perform association rule mining in various retail datasets

Shipment price prediction

Developed ML model to predict shipment price in logistics of a pharma company with an accuracy of 95%

More projects on Github

I love to solve business problems & uncover hidden data stories


GitHub

Contact

Contact Me

Below are the details to reach out to me!

Address

Newark, NJ

Contact Number

+ 1 (862) 405-2014

Email Address

tanmaysk22@gmail.com

Download Resume

Download ⬇️



Have a Question? Click Here


  • Find me on