ElleHacks 2026 AI Document Analysis Platform

An AI-powered document explainer prototype that helps users understand legal, government, and financial documents through age-adaptive explanations and retrieval-based document analysis.

What the project does

The current working demo focuses on the document upload workflow. Users can upload a typed PDF or paste document text, prepare the document as searchable embeddings, and generate a simplified explanation from saved or retrieved document chunks.

Embeddings-first RAG workflow

01

Upload or Paste

The user uploads a typed PDF or pastes document text into the web app.

02

Extract Text

PDF.js extracts readable text from the uploaded document in the browser.

03

Chunk + Embed

The backend chunks the extracted text and creates embeddings for semantic retrieval.

04

Save to MongoDB

Searchable chunks and embeddings are stored in MongoDB Atlas Vector Search.

05

Retrieve Context

The app searches saved chunks to retrieve the most relevant document context.

06

Generate Explanation

Gemini generates an age-adaptive explanation using the retrieved document context.

What I worked on

Built the document upload and paste-text interface.

Connected PDF text extraction using PDF.js.

Implemented age-adaptive explanation logic with Gemini AI.

Added MongoDB Atlas Vector Search for semantic document retrieval.

Updated the flow to index document chunks before generation for a more privacy-conscious RAG workflow.

Created backend API documentation for testing the document workflow.

How the demo handles document context

The demo does not permanently store the uploaded PDF file itself. Instead, it extracts text, saves searchable chunks and embeddings, and generates explanations from saved or retrieved context. For a production system, the next step would be adding automatic redaction, user-level access controls, and document retention settings before storing sensitive client content.

Tools used

Next.js TypeScript Gemini AI MongoDB Atlas Vector Search PDF.js Vercel RAG Semantic Search

Try the working document upload workflow.

The demo is protected because it connects to live AI and database services.

Launch Protected Demo