- 👋 Hi, I’m Harsh Belekar — an aspiring AI & Data Analytics Engineer passionate about turning raw data into meaningful insights and building intelligent systems.
- 🧠 Skilled in SQL, Power BI, Tableau, Python, and Excel, with hands-on experience in projects like Swiggy Sales Analysis, Samsung Data Engineering, and other real-world data projects.
- 🎯 Actively exploring the fields of AI/ML, Data Science, and Data Analytics to strengthen my problem-solving and analytical capabilities.
- 🤖 I have built an Autonomous Drone using Raspberry Pi Zero W, integrating sensor fusion, computer vision, and automation.
- 📊 Regularly work on Data analysis, ETL workflows, Dashboards, and predictive modeling across SQL, Power BI, Excel, and Python ecosystems.
- 🎮 Develop games like Shinobi Saga, combining programming, UI/UX, and creative design.
- 💡 Passionate about learning new technologies, solving complex problems, and creating meaningful projects that make an impact.
A 2D Naruto vs Sasuke fighting game made with Pygame, featuring animations, health bars, weapons, and special attacks.
An End-to-End production-grade analytics project analysing 600,000+ food delivery orders across 10 Indian cities (2022–2025) using Python, PostgreSQL, and Power BI. Covers the complete pipeline — synthetic data generation, star schema design, ETL loading, advanced Python EDA, and a 7-page interactive Power BI dashboard — solving 7 real business challenges including customer retention, revenue optimisation, restaurant performance, and market expansion strategy.
This project builds a production-grade Data Warehouse for Samsung — from scratch — using the Medallion Architecture (Bronze → Silver → Gold). The pipeline begins with Synthetic Data Generation simulating Samsung India's real business operations across Sales, Finance, Customer Relations, After-Sales Service, Supply Chain, and Marketing, then processes that data through three structured warehouse layers into analytics-ready Gold views.
A comprehensive End-to-End SQL Data Warehouse project built in PostgreSQL, implementing a full Bronze → Silver → Gold layer architecture. Integrates ERP and CRM source data, builds a Star Schema with clean dimension and fact tables, develops ETL pipelines, and applies 12 modules of Advanced SQL Analytics — covering customer segmentation, product performance, cohort analysis, and cumulative reporting.
A Bluetooth-controlled 4WD Robotic Arm Vehicle built using Arduino Uno, featuring a 4-DOF Robotic Arm, Voice Command Support, and a Custom Android Application developed with MIT App Inventor. The robot combines a powerful 4-wheel-drive rover base with a servo-controlled robotic arm, allowing users to remotely navigate the vehicle and manipulate objects wirelessly.
A fully Interactive Excel dashboard analysing Blinkit's grocery sales performance using KPIs, pivot charts, and dynamic slicers. Covers outlet-wise sales breakdown, item-category analysis, fat-content segmentation, and customer rating trends — delivering retail analytics insights entirely within Excel without any external tools.
⭐ “Learning never stops when curiosity drives creation.” ⭐