ClearDOCSAI

Inspiration

Watching people panic over medical and insurance documents they couldn't understand made us want to build something that brings calm and clarity to one of the most stressful everyday experiences.

What it does

ClearDocs AI explains complex medical, insurance, and legal documents in plain English using a fully local LLM — highlighting unclear phrases, flagging alarming language, and suggesting questions to ask a professional.

How we built it

We built a Python Flask backend that connects to Ollama running llama3 locally, with a minimal HTML/JavaScript frontend served from the same server — no cloud, no API keys, one command to run everything.

Challenges we ran into

Getting a consistent structured JSON response from the local LLM without hallucinations or broken formatting took significant prompt engineering and fallback handling.

Accomplishments that we're proud of

We built a fully functional, privacy-first AI tool that runs entirely on a local machine and genuinely makes intimidating documents feel approachable — with zero external data exposure.

What we learned

Local LLMs are powerful enough for real-world NLP tasks when paired with carefully designed prompts, and privacy-first architecture is not just possible but practical even in a hackathon timeframe.

What's next for ClearDOCSAI

We plan to add PDF and DOCX upload support, auto document-type detection, multilingual explanations, and a confidence scoring system to flag when the AI is uncertain about its interpretation.

Built With

• and-python-dotenv-?-all-running-locally-with-no-cloud-services
• databases
• external
• flask
• html
• javascript
• ollama-(llama3)
• or
• python