I design and ship ML systems, full-stack web apps, and freelance production work — from a pest-outbreak model published at ICRTSET-2026 to a lead-gen agency site live in production.
Computer Engineering graduate (CGPA 7.44/10) from SPPU with hands-on experience across Python, Flask, machine learning, and full-stack web development. I published research on ML-based pest outbreak prediction for Kharif crops (ICRTSET-2026), and I run freelance/agency-style production work — building animated content channels and client websites end-to-end, under real deadlines and real client constraints.
Languages, frameworks, and tools I build with day to day.
Three builds spanning applied ML, production web apps, and full-stack fundamentals.
An ML system predicting pest outbreaks across Maharashtra districts using environmental and soil data from 1,000+ sensor records. Built a data pipeline (outlier detection, feature encoding, normalization) that improved model input quality by ~40%, and trained/compared Random Forest, XGBoost, SVM, and LSTM models. Includes a Flask backend for real-time predictions and a dashboard visualizing pest risk and weather trends. Research accepted at ICRTSET-2026.
A high-performance, lead-focused agency website built for HVAC & Solar businesses, with a component-driven React/TypeScript/Tailwind UI. Structured the codebase (API layer, reusable components, hooks) for maintainability, and deployed the production build via Vercel — built collaboratively with a friend from design through deployment.
A full-stack CRUD web app to manage student records (add, view, delete) with a clean HTML/CSS interface. Implemented Flask-SQLAlchemy ORM with SQLite, structured for deployment with a requirements.txt and documented on GitHub with a clear README for reproducibility.
Built and tested Python scripts and Flask web applications for real internship project requirements. Cleaned and processed datasets using Pandas and NumPy, wrote modular PEP-8 compliant code, and collaborated via Git with a team of 5 across weekly sprints.
Evaluate and rate Google Maps search results for relevance, accuracy, and quality using detailed evaluation guidelines.
Perform structured evaluation of AI model outputs for data quality, accuracy, and alignment with project guidelines.
Presented at the International Conference on Recent Trends in Science, Engineering & Technology. Explores environmental feature-based classification models — Random Forest, XGBoost, SVM, and LSTM — for agricultural pest risk prediction across Maharashtra.