Intelligent Onboarding Automation

This Django-based application automates the onboarding process for MagicBus foundation's skilling and job placement programs. It targets candidates aged 18-25 who have passed 12th grade, have a family income less than 3 lakh per annum, and are not currently working.

Features

• Candidate registration and eligibility validation
• Document upload (Aadhaar, 12th Marksheet, Income Certificate) - individual or bulk
• OCR extraction using Azure Cognitive Services
• Fraud detection for duplicate/fake documents
• Interactive React dashboard with metrics and management
• REST API for integration

Setup

1. Install dependencies:

   pip install -r requirements.txt
   

2. Configure Azure services:

   • Set up Azure Computer Vision for OCR
   • Update settings with Azure credentials

3. Run migrations:

   python manage.py migrate
   

4. Start the server:

   python manage.py runserver
   

5. Run the React dashboard:

   cd frontend
   npm install
   npm run dev
   

API Endpoints

Candidates

• GET /api/candidates/ - List all candidates
• POST /api/candidates/ - Create a new candidate
• GET /api/candidates/{id}/ - Retrieve a candidate
• PUT /api/candidates/{id}/ - Update a candidate
• DELETE /api/candidates/{id}/ - Delete a candidate
• POST /api/candidates/{id}/validate_eligibility/ - Validate eligibility
• POST /api/candidates/{id}/upload_documents/ - Upload multiple documents (Aadhaar, 12th Marksheet, Income Certificate)

Documents

• GET /api/documents/ - List all documents
• POST /api/documents/ - Upload a new document
• GET /api/documents/{id}/ - Retrieve a document
• PUT /api/documents/{id}/ - Update a document
• DELETE /api/documents/{id}/ - Delete a document

Database Schema

Candidate Table

• id: Primary Key
• name: CharField
• age: IntegerField (18-25)
• email: EmailField (unique)
• phone: CharField
• address: TextField
• twelfth_pass: BooleanField
• family_income: DecimalField (max 300000)
• not_working: BooleanField
• created_at: DateTimeField
• eligible: BooleanField

Document Table

• id: Primary Key
• candidate: ForeignKey to Candidate
• document_type: CharField (aadhar, Twelfth, income)
• file: FileField
• ocr_text: TextField
• verified: BooleanField
• hash_value: CharField (for fraud detection)
• uploaded_at: DateTimeField

Eligibility Criteria

• Age: 18-25
• 12th Pass: Yes
• Family Income: <= 3,00,000 INR
• Employment Status: Not working

Azure Integration

• OCR: Azure Computer Vision API
• Authentication: Azure Identity

React Dashboard

The project includes an interactive dashboard built with React and Vite for easy management:

• Dashboard: View key metrics, charts, and recent candidates
• Candidates: List, view details, and validate eligibility
• Add Candidate: Form to create candidates with document uploads
• Documents: View and filter documents by candidate

Run the dashboard with: cd frontend && npm run dev

The dashboard connects to the Django API at http://localhost:8000/api.

Import the onboarding_api_insomnia.json file into Insomnia to test the API endpoints. Set the environment variable base_url to http://localhost:8000/api and candidate_id/document_id as needed for specific requests.

Future Enhancements

• Frontend portal using Streamlit
• AI-powered recommendations using LangChain
• Data processing with Databricks