EDUCATION
-
Vellore Institute of Technology, Vellore
M.Sc. in Data Science | 8.79 CGPA 08/2023-07/2025
-
Maulana Abul Kalam Azad University of Technology, West Bengal
B.Sc. in Data Science | 9.44 CGPA 09/2020-08/2023
-
Kalyani Public School, Barasat
CBSE | Class XII | 70.8% 04/2018-03/2020
-
The Central Modern School, Barasat
ICSE | Class X | 78.6% 04/2008-03/2018
SKILLS
- Programming Languages: SQL, Python(Pandas, NumPy, Matplotlib, Statsmodels, Seaborn, Keras, Scikit-learn, scipy, Tensorflow, PyTorch, NLTK, re, Transformers),
R(BSDA, ggplot2, dplyr, ggthemes, readxl, tinytex, knitr), Java.
- Tools: Oracle, MySQL, Tableau, Power BI, Excel, Jupyter Notebook, Google Colab, AWS(Sagemaker), RStudio, Visual Studio Code, Git,
Github, Overleaf, ChatGPT, Claude, Adobe Photoshop, Adobe Illustrator.
- Technical: Data Analysis and Visualization, Business Intelligence, ETL processes, Machine Learning, Deep Learning, Image Processing, Natural Language Processing,
Time Series Analysis, Regression Analysis, Fine Tuning LLMs, Prompt Engineering.
WORK EXPERIENCE
-
Machine Learning Intern | Cognifyz Technologies 05/2024-06/2024
- Developed a regression model with a R-squared value of 95.24% for a restaurant recommendation system.
- Performed geospatial analysis to understand customer behavior, improving customer engagement by 20%.
-
Data Analyst Intern | Edulyt India 01/2024-03/2024
- Cleaned 100+ rows from a credit banking dataset and extracted customer monthly spend, repay, and due amounts using Python.
- Analyzed credit card transactions to identify top spending cities, trends by card type, spending ratios, and gender-based contributions using SQL.
- Cleaned 50+ rows of missing credit card data, ensured transaction accuracy, and segmented customers by spend patterns, conducting detailed spend and
return analysis across categories using Python.
-
Data Science Intern | Let’sGrowMore 01/2024-02/2024
- Prepared an interactive dashboard using Power Bi on Terrorism dataset, where Middle East and North Africa has highest number of attacks(50,000) for the last 20 years.
- Developed an Iterative Dichotomiser 3 (ID3) decision tree from scratch without using pre-existing libraries which gave accuracy of 100% over the iris dataset.
- Implemented a predictive language model using RNN which achieved an accuracy of 87%.
-
Data Science Intern | ThinkAgainLab 11/2022-02/2023
- Completed projects in descriptive and exploratory data analysis on Uber Dataset using R discovered a 175% increase in NY Uber trips between 5 PM to 7 PM and in summer season.
- Executed a comprehensive analysis of HR analysis data to find whether there is any correlation between education level and job satisfaction, revealing a correlation value of-0.01.
PROJECTS
-
PyroAlert (Ongoing)
- Developing an AI-powered fire detection system utilizing CNN for real-time image and video analysis, identifying and localizing fire hazards.
-
Potato Disease Classification Using VGG16
- Developed a potato disease classification model using the VGG16 architecture, processing a dataset of potato leaf images to identify diseases like Early Blight and Pink Rot.
The dataset was pre-processed through resizing, image inversion, and data augmentation techniques, generating over 2500 images per class.
- Evaluated the model’s performance using precision 97.94%, recall 97.89%, F1-score 97.88%, test accuracy 97.89%, and confusion matrix, along with ROC and
Precision-Recall curves for individual classes.
-
State-wise Business Comparison and Forecasting
- The data was collected from the Ministry of Corporate Affairs, Government of India, as part of the Open Government Data (OGD) Platform, which provides access to
various datasets for public use.
- Analyzed 28 states of India to find the top 5 principal business activities and liquidity ratio.
- Forecasted the growth of the principal business activities over the next 5 years along with comparative analysis across different states.
Designed and deployed a web application using Streamlit, which assists investors.
-
Generalized Medicine Recommendation System
- Collected data from 150+ patients through surveys, cleaned and pre-processed data, and performed a comprehensive analysis to assess data integrity and distribution.
- Implemented a decision tree classifier for disease classification and medication suggestion, achieving 83% accuracy, 75% precision, and 83% recall.
EXTRACURRICULAR ACTIVITIES
- Volunteered at the International Conference on Data Management, increasing engagement by 50%.
- Designed and executed graphic materials for multiple events at VIT.