Data Analyst • Data Scientist • ML Builder
Final-year Computer Engineering student at Parul University, Vadodara. I turn messy data into decisions - cleaning it, finding the story in it, and shipping the result as a dashboard or a live app.
| Project | What it does | Stack |
|---|---|---|
| Customer Churn Prediction | Predicts telecom churn - 79.84% accuracy, live Streamlit app | Python, Scikit-learn, XGBoost, Streamlit |
| E-commerce Sales Analysis | Found $100K+ losses across 9,994 orders | Python, SQL, Power BI |
| COVID-19 India EDA | State-wise analysis of 36 states with Power BI dashboard | Python, Plotly, Power BI, SQL |
| A/B Testing Analysis | Statistical test showed new landing page should not launch | Python, SciPy, StatsModels |
| HR Attrition Analysis | Predicts which employees will leave by department, age and role | Python, Scikit-learn, Power BI |
| Project | What it does | Stack |
|---|---|---|
| KrishiMitra+ | Crop disease detection at ~98% accuracy + farmer marketplace | React, TensorFlow, Supabase, Hugging Face |
| ExpenseAI | AI finance app with UPI tracking and Claude API advisor | React, Node.js, Claude API |
| WhosMyMate | Roommate matching with weighted compatibility scores, AI explanations and messaging | TypeScript, Supabase, Claude API |
| Tripzy | Ride-hailing app with AI chatbot, fare prediction and driver anomaly detection | React, Node.js, ML |
Data Analysis
pandas NumPy EDA Statistics A/B Testing Excel
Machine Learning
Scikit-learn XGBoost Logistic Regression Random Forest TensorFlow
Visualization
Power BI Matplotlib Seaborn Plotly
Databases
SQL MySQL PostgreSQL Window Functions
Deployment
Streamlit Vercel Netlify GitHub
Web Development
React Node.js Express Supabase TypeScript
