FormSaathi AI 把政府通告整理成清晰的行動步驟
FormSaathi AI 🇮🇳: Turning Government Notices into Clear Next Steps
FormSaathi AI 是一款協助用戶理解政府通告、獎學金公告及公共服務通函的網頁應用程式，可貼上文字或上載圖片，並以 OCR 提取內容，整理重點、日期、所需文件及後續步驟。應用程式支援英文和印地語，已部署首個版本；開放權重模型的即時推理仍在除錯，尚未驗證。
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady
What I Built
I built FormSaathi AI 🇮🇳, a web application that helps people understand government notices, scholarship announcements, and public-service circulars without manually digging through long paragraphs.
Users can paste notice text or upload an image. The application extracts text using OCR and organizes information into important points, dates, document mentions, and practical next steps. It supports both English and Hindi.
The goal is to reduce the effort required to understand complicated notices and make important information easier to find.
Demo
🚀 Live Demo: https://formsaathi-ai-user.vercel.app/
The frontend and backend are deployed on Vercel.
Code
💻 GitHub Repository: https://github.com/Shivam-Sharma2009/FormSaathi-AI
The repository contains the frontend, Express backend, OCR functionality, and AI-provider integration.
How I Built It
I built FormSaathi AI using:
- Frontend: HTML, CSS, and JavaScript
- Backend: Node.js and Express.js
- OCR: Tesseract.js
- Deployment: Vercel
- AI inference integration: Hugging Face Inference Providers with an OpenAI-compatible API interface
I wanted to develop the project without downloading large language models onto my laptop, so I chose a cloud-based workflow.
The intended AI workflow is to summarize notices, identify important dates and document requirements, and produce a numbered action checklist. The application also checks supporting evidence against the supplied notice text to help users verify the extracted details.
The open-weight model integration is currently being debugged, and live inference has not yet been verified. The deployed application uses a rule-based fallback when the inference provider is unavailable.
Why Does Open Innovation Matter?
Open innovation gives students and independent developers more opportunities to experiment with AI without requiring expensive hardware or building everything from scratch.
For FormSaathi AI, open-weight models offer the possibility of greater flexibility in model selection, inference providers, and future deployment choices.
I want to keep the project adaptable rather than permanently tied to one AI vendor. Open ecosystems make it easier to learn, experiment, and improve a project as it grows.
My Agent Session
I haven't published a DevRelay agent session for this project yet. I hope to add one as development continues.
The Lazy Payoff
Instead of manually scanning a long notice to locate dates, document requirements, and important instructions, users can use FormSaathi AI to organize the information in one place.
The first deployed version is available. My next goal is to get live open-weight inference working reliably and measure the time saved through real-world testing.
Prize Categories
I will verify the partner-specific eligibility requirements before claiming any prize category.
Built by a student developer working to make complicated public information easier to understand. 🇮🇳
Feedback and contributions are welcome!
來源：DEV Community · dev.to