An intelligent web application designed to analyze and provide detailed feedback on proposals. This tool leverages the Cerebras API with Llama models to evaluate proposals based on key metrics, identify strengths and weaknesses, and offer constructive advice. The entire application is containerized with Docker for easy deployment and use.
- PDF Proposal Upload: Easily upload proposals in PDF format.
- AI-Powered Metrics Analysis: Get scores on crucial metrics like Technical Depth, Project Understanding, and Timeline Clarity.
- Strengths & Weaknesses: The AI identifies the top three strengths and areas for improvement in the proposal.
- Timeline Extraction: Automatically extracts the proposed project timeline from the document.
- Detailed Constructive Feedback: Generates a comprehensive, prose-based review to help applicants improve their proposals.
- Containerized & Portable: Packaged with Docker for a seamless setup on any machine.
This project is distributed as a Docker container. You can run it on any machine with Docker installed without needing to set up a development environment.
Pull the latest image from the GitHub Container Registry:
docker pull ghcr.io/ikartiksh/gsoc-proposal-reviewer:latestFrontend: React, Vite
Core Backend: Node.js
AI Backend: Cerebras API (Llama Models)
PDF Processing: pdf.js
Deployment: Docker, Nginx, GitHub Container Registry