My Projects
AI-Powered CV Analysis Assistant
An AI-powered tool to upload and analyze CVs using OpenAI's models. It extracts insights and answers natural language questions, helping recruiters quickly screen candidates through a simple Streamlit interface.
React + FastAPI OpenAI Prompt App
A full-stack application that allows users to input prompts via a React frontend, sends them to a FastAPI backend, and fetches intelligent responses from OpenAI's GPT model. The system features asynchronous handling, CORS configuration, and uses environment variables for secure API access.
Amazon Review Star Rating Predictor using Tkinter and Machine Learning
A tkinter based project that allows using Tkinter that loads a trained sentiment rating model and vectorizer to predict Amazon-style review ratings (out of 5 stars) based on user-entered review text.
Streamlit project on Employee Attrition Prediction System
This Streamlit app predicts employee attrition based on various input features using a pre-trained machine learning model, with data preprocessing, model inference, and result visualization.
Tkinter project on Email Spam Detection
This code creates a simple email spam detection application with a graphical user interface (GUI) using Tkinter, where users can input email content, and the model predicts whether the message is spam or not.
Temple Info Chatbot with RAG using Gemini
This app is a Streamlit-based chatbot that uses Retrieval-Augmented Generation (RAG) to answer questions about temples in Nepal using embedded PDF data. It combines Google’s Gemini LLM with FAISS for semantic search to provide contextual, detailed responses.
Streamlit-Based Sentiment Analysis App for Musical Instrument Reviews
This Python script uses Streamlit to build an interactive web application that loads a pre-trained multilayer perceptron (MLP) sentiment analysis model and a CountVectorizer, allowing users to input or select musical instrument reviews and receive sentiment predictions (Positive, Negative, or Neutral) along with visual feedback using emojis.
Insurance Charges Predictor Using Linear Regression and Streamlit
This Streamlit app trains a Linear Regression model on a preprocessed insurance dataset to predict medical charges based on user input (age, BMI, smoker status, region, and number of children) and displays both the predicted charges and model performance metrics.
NTC Stock Price Analysis and Prediction Using Linear Regression in Streamlit
This Streamlit app visualizes historical NTC stock data and predicts future closing prices using a linear regression model based on open, high, and low prices.
Nabil Bank Stock Price Analysis and Prediction Using Linear Regression in Streamlit
This Streamlit app visualizes historical Nabil Bank stock data and predicts future closing prices using a linear regression model based on open, high, and low prices.
FastAPI-based Chatbot with Google Gemini Integration
This FastAPI application interacts with Google Gemini's LLM to generate responses based on user-provided prompts. It exposes a simple POST endpoint where users can send prompts and receive text-based answers from the Gemini model.