DocSimplified

Landing Page
Document Analysis
Registration
Document Dashboard

Inspiration

Government forms are notoriously confusing — even for families who have lived in the U.S. for decades. Three out of four of our team members come from immigrant households, and we’ve all watched our parents struggle with taxes, visa renewals, and federal documentation because of language barriers, dense terminology, and inaccessible formatting. We wanted to build a tool that makes essential government services understandable for everyone, regardless of literacy level or language proficiency.

What it does

DocSimplified is a multilingual AI-powered web application that simplifies government documents. Users upload any fillable PDF, and the app automatically detects form fields, summarizes each section in clear steps, translates the content into their preferred language, and guides them through filling it out. It even includes text-to-speech for users with low vision or reading difficulties. The goal: make complex forms feel intuitive and approachable.

How we built it

Frontend: React and Tailwind for a clean, accessible UI; pdf-lib for rendering and editing fillable PDFs.
Backend: Supabase for authentication, user profiles, and document storage.
AI Processing: OpenAI 4o API for summarization, language translation, and adaptive explanations tuned to the user’s education level.
Accessibility: ElevenLabs for high-quality text-to-speech output in multiple languages. We tied all of these components together to create a smooth workflow from upload → analysis → guided form-filling → saving progress.

Challenges we ran into

Our initial PDF viewer couldn’t read embedded form fields, forcing us to switch to a more reliable parsing method.
OCR-based image uploads were too time-intensive to implement during the hackathon.
Early OpenCV bounding-box detection only recognized perfect rectangles, while many real forms used underlines or irregular shapes.
Integrating multiple third-party APIs while keeping the interface fast and accessible required lots of debugging and design iteration.

Accomplishments that we're proud of

Delivering a fully functional, end-to-end document simplification and form-filling experience.
Building a genuinely accessible tool with translation, easy summaries, and text-to-speech baked in.
Creating a polished and intuitive UI despite the technical complexity underneath.
Staying true to our mission of helping immigrant and low-literacy communities — the people who inspired this project in the first place.