Developed a movie recommendation system using cosine similarity and Count Vectorizer for text representation. Created an interactive web app with Streamlit for real-time recommendations, enabling users to explore movie similarities. The scalable system leverages advanced NLP techniques for personalized suggestions.

Developed a house price prediction model using AdaBoost, deployed via Streamlit for user-friendly access. The project included data preprocessing, feature engineering, model evaluation, and tuning. Plans for scalability and advanced algorithms aim to enhance predictions for real estate professionals and buyers.

Developed a used car price prediction model utilizing machine learning algorithms like Linear Regression and Random Forest. Created an interactive Streamlit app for instant price predictions based on car features. The project aims to aid informed buying and selling decisions in the market.

Developed a WooCommerce e-commerce solution with a custom theme, integrated payment gateways, and automated inventory management. Focused on SEO and performance optimization, while implementing analytics for data-driven insights, ultimately enhancing user experience and boosting online sales and customer satisfaction.

Developed a job email generation system leveraging the LLaMA language model, LangChain for seamless integration, and Streamlit for an interactive front-end interface. The system generates personalized job application emails based on user input, streamlining the application process.

Developed a potato leaf disease prediction system using deep learning techniques to accurately identify and classify diseases based on images of potato leaves. The deep learning model was trained on a large dataset of labeled images, enabling high accuracy in disease detection. The system was hosted using FastAPI, providing a fast and efficient backend to process user inputs and return predictions. The front-end interface was built with ReactJS

Developed a static website using HTML,CSS,JS,PHP for the client who owns a travel agency and cab service business.

This project implements Twitter Sentiment Analysis using a **BERT (bert-base-uncased) model trained ** with PyTorch. The model classifies tweets into positive, negative, and neutral sentiments.

Developed a woocommerce website for a client who wanted to sell his/her medical products online.

Woked on a research paper on differnet fraud detection techniques using mix of ML and DL models.