A rebuilt EPFO portal that runs automatic pre-validation checks before any PF claim is submitted, tells the member exactly what will cause rejection, and generates the specific document needed to fix it. Built for the Build What Moves India 2026 Hackathon.
| Feature | Description |
|---|---|
| Smart Pre-validation | Runs 4 automated checks (Name Match, DOB, Exit Date, Bank KYC) before submission, intercepting common rejection causes instantly. |
| AI Name Matching | Uses GPT-4.5 fuzzy matching to compare EPF records against Aadhaar, accommodating minor spelling variations that usually trigger manual rejection. |
| Auto-Document Generation | Automatically drafts required legal documents (e.g., Joint Declaration) using jsPDF based on the specific failure reason. |
| Actionable Resolution | Replaces generic rejection codes with plain-language explanations, numbered fix steps, and direct links to official EPFiGMS portals. |
| Accessible & Bilingual | Modern, mobile-first design with English and Hindi support, completely replacing the clunky government interface. |
| Secure Architecture | Client-side execution with sensitive AI operations routed securely through Vercel Serverless Functions. |
The Problem: In 2024-25, 174 lakh PF claims were rejected — 1 in every 5. The current portal rejects with a generic code (like "Name Mismatch"). No explanation. No fix path. Workers are forced to visit EPFO offices multiple times, often losing wages just to figure out why they were rejected.
The Solution: A pre-validation layer inserted before submission. Instead of learning about a problem after rejection, members see exactly what is wrong and how to fix it in seconds.
-
Clone the repository and install dependencies:
git clone https://github.com/h55n/epfo-validator.git cd epfo-validator npm install -
Configure Environment:
cp .env.example .env.local
Add your
VITE_AI_PROVIDERandAI_API_KEYto.env.local. -
Start the development server:
npm run dev
Use these mock accounts to test different validation scenarios on the live app or locally:
| Account Name | UAN Number | Password | Expected Scenario |
|---|---|---|---|
| Ramesh Kumar | 100673291847 | Demo@1234 | Name mismatch (AI validation) |
| Fatima Shaikh | 100891234567 | Demo@1234 | Multiple failures (Exit Date) |
| Vijay Patil | 100334455678 | Demo@1234 | All clear (Happy path) |
Divided and developed equally by:
- MRUNMAYEE DAWARE
- HASSAN REHMAN
Disclaimer: This is an independent hackathon prototype and is not affiliated with the Employees' Provident Fund Organisation (EPFO) or the Government of India.