Projects

Projects

Selected AI projects and additional web development work with practical outcomes.

Featured Projects

A selection of my recent work in Machine Learning and modern web development.

Machine Learning-Based Personal Finance Assistant dashboard preview

Machine Learning-Based Personal Finance Assistant

A Final year academic project that helps manage personal finances more intelligently and accessibly using modern web development and machine learning.

  • Automatically categorizes transactions with a Logistic Regression model trained on a large transactions dataset.
  • Predicts next month's spending per category using Linear Regression trained on each user's transaction history.
  • Detects unusual transactions in real time with Isolation Forest and sends immediate notifications.
  • Tracks budgets with automated warnings, spending analytics, recurring transaction management, and multi-format export.

Impact & OutcomeBuilt to make personal finance management smarter, faster, and easier to access.

Personal FinanceMachine LearningNext.jsFastAPIPostgreSQL
Movie recommender system dashboard preview

Movie Recommender System

Movie recommendation web app using the TMDB dataset and classical machine learning techniques.

  • Used NumPy, Pandas, and NLTK for data preprocessing.
  • Used scikit-learn to build recommendation logic.
  • Built and deployed a user-facing demo application.

Impact & OutcomeProvides fast and simple movie recommendations based on user interest.

Recommendation SystemNLPStreamlitDeployment
RAG AI teaching assistant interface preview

RAG AI Teaching Assistant

RAG-based teaching assistant that answers questions from video-derived learning content.

  • Built a context-aware assistant for video-trained knowledge.
  • Stack included OpenAI Whisper, BGE-M3, scikit-learn, and Llama 3.2.

Impact & OutcomeHelps users find relevant learning points from video content faster.

RAGLLMEmbeddingsSpeech-to-Text
Explore all other projects

Other Web Development Work

  • Built full stack web applications using MERN and Next.js.
  • Delivered WordPress websites used by well-known companies in Nepal.
  • Handled end-to-end work from UI to backend and deployment.