projects

Here are some of the projects I've been working on:

Major Projects

Boltly

WebContainer APITypeScriptClaude AI API
  • Utilized the WebContainer API to enable an in-browser Node.js environment for a seamless development experience.
  • Integrated Claude AI API for dynamic content generation and enhanced interactivity.
  • Developed using TypeScript for robust type safety and maintainability.
  • Implemented a code editor using Monaco Editor, allowing users to write and test code directly in the browser due to web containers.
  • Created a responsive UI with Vite-powered development and Tailwind CSS for modern styling.
GitHub

Sync Script

Mongo DBSocket ioNodeReactExpress
  • Developed Sync-Script, a real-time collaborative text editor supporting multiple users simultaneously.
  • Implemented instant synchronization of document changes across clients using Web-Sockets.
  • Configured automatic periodic saving to prevent data loss during editing sessions.
  • Designed an intuitive interface with customizable toolbar options for enhanced user experience.
  • Enabled seamless document management through unique identifiers for creating and accessing documents.
GitHubLive Demo

AI Search Agent (Perplexity Clone)

GroqGoogle Search APINodeReactExpress
  • Leveraged Llama-3.3 AI for natural language understanding and processing.
  • Integrated Google Custom Search API to provide reliable web search results.
  • Created a clean, responsive UI to ensure seamless user experience across devices.
GitHubLive Demo

Justice-GPT

Gemini AI APIReact
  • Developed Gemini AI Crime Case Generator, a web-based tool to analyze crime cases based on the Indian Constitution.
  • Integrates the Gemini API to map user-submitted crime details to relevant legal articles and case analysis.
  • Features a responsive UI built with React, TypeScript, and Tailwind CSS for a seamless user experience across devices.
GitHubLive Demo

Micro-GPT

PythonJupyter NotebookEmbeddingsAttention
  • Implemented "Attention Is All You Need" architecture in Python for local GPT-style query handling.
  • Designed a lightweight model optimized for offline text embeddings and question-answering.
  • Prioritized data privacy, faster response times, and control in local environments.
GitHub