AI-powered funding discovery platform connecting Young, Beginning, and Small (YBS) farmers with federal grants, loans, and subsidies. Click here to visit Cultura live.
This is a functional prototype built for the AgFirst/SCSU Hackathon (February 20-22, 2026). While the core features are implemented and demonstrated, this is not production-ready software. It showcases the concept, architecture, and potential impact of an AI-powered funding discovery platform for farmers.
Current Status:
- ✅ Functional demo with sample data
- ✅ Core features implemented
- ✅ Architecture designed for scale
⚠️ Not connected to live USDA systems⚠️ Sample data for demonstration purposes⚠️ Requires additional testing and security hardening for production use
Interested in taking this further? See Contact below!
60% of eligible farmers never apply for USDA funding because:
- Navigating 2,000+ programs is overwhelming
- Research takes 15+ hours per application
- Eligibility requirements are unclear
- Deadlines are easy to miss
Result: $2 billion in federal funding goes unclaimed annually.
Cultura simplifies funding discovery with:
- 🎴 Swipe-Based Discovery - Browse opportunities with AI-powered match scores
- 🤖 Smart Matching - AI analyzes your farm profile against 2,000+ USDA programs
- 📊 Unified Dashboard - Track all applications with color-coded status indicators
- 📝 Guided Applications - Step-by-step wizards with pre-filled forms
- 🔔 Smart Notifications - Never miss a deadline or status update
- 15-minute profile creation (vs. hours of research)
- 5-minute opportunity discovery (vs. weeks of searching)
- Real-time matching with 2TB+ USAspending.gov database
- Mobile-first design - Apply from anywhere
- Offline support - Works with limited connectivity
- WCAG 2.1 AA accessible - Inclusive for all farmers
cultura/
├── frontend/ # React 18 + TypeScript + Vite
├── backend/ # AWS Lambda functions (Node.js 20)
├── infrastructure/ # AWS CDK infrastructure as code
├── shared/ # Shared TypeScript types & validation
└── docs/ # Comprehensive documentation
- Node.js 20.x or higher
- npm 10.x or higher
- AWS CLI configured with credentials
- AWS CDK CLI:
npm install -g aws-cdk
# Clone the repository
git clone https://github.com/dajley/Cultura.git
cd cultura
# Install dependencies for all packages
npm install# Start frontend development server
cd frontend
npm run dev
# Visit http://localhost:3000
# Run tests
npm run test
# Build all packages
npm run build# Bootstrap CDK (first time only)
cd infrastructure
npx cdk bootstrap
# Deploy infrastructure
npm run deploy
# Deploy backend services
cd ../backend
npm run deploy
# Deploy frontend
cd ../frontend
npm run deploySee DEPLOYMENT.md for detailed deployment instructions.
Copy .env.example to .env and configure:
# AWS Configuration
AWS_REGION=us-east-1
AWS_ACCOUNT_ID=your-account-id
# Database
POSTGRES_HOST=your-db-host
POSTGRES_DB=usaspending
POSTGRES_USER=your-user
POSTGRES_PASSWORD=your-password
# API & Auth
VITE_API_URL=https://api.cultura.com
VITE_USER_POOL_ID=us-east-1_xxxxx
VITE_USER_POOL_CLIENT_ID=xxxxxSee docs/ENVIRONMENT_VARIABLES.md for all configuration options.
- React 18 - Modern UI library
- TypeScript - Type-safe development
- Vite - Lightning-fast build tool
- Tailwind CSS - Utility-first styling
- React Router - Client-side routing
- AWS Amplify - Authentication & API
- AWS Lambda - Serverless compute (Node.js 20)
- API Gateway - RESTful API endpoints
- DynamoDB - NoSQL database for user data
- PostgreSQL - USAspending.gov data (2TB+)
- Redis (ElastiCache) - Caching layer
- AWS Bedrock (Claude 3) - AI/ML matching
- AWS CDK - Infrastructure as code
- CloudFront - Global CDN
- S3 - Static asset hosting
- Cognito - User authentication
- CloudWatch - Monitoring & logging
- EventBridge - Scheduled notifications
- Vitest - Unit & integration tests
- fast-check - Property-based testing
- React Testing Library - Component tests
- ⚡ < 3 second page loads
- 🚀 < 1 second database queries (with caching)
- 📱 Mobile-optimized for rural connectivity
- 💾 Offline support via service workers
- 🎯 95%+ match accuracy with AI recommendations
-
Create Profile (15 minutes)
- Farm details, crops, location
- YBS status auto-calculated
-
Discover Matches (5 minutes)
- Swipe through AI-ranked opportunities
- See match scores and explanations
-
Apply with Guidance
- Pre-filled application forms
- Document upload support
- Progress tracking
-
Track Applications
- Color-coded status dashboard
- Deadline reminders
- Funding pipeline overview
- Requirements - Detailed requirements
- Design Document - Architecture & design
- Deployment Guide - Step-by-step deployment
- Demo Script - Presentation guide
- Troubleshooting - Common issues & solutions
- Performance Optimizations - Speed improvements
This is a hackathon prototype built to demonstrate the concept. Feel free to open an issue if you have questions about the implementation or ideas for improvement!
Target Metrics:
- 📊 Increase application rates from 40% to 80%
- ⏱️ Reduce research time from 15 hours to 15 minutes
- 💰 Help farmers access $2B in unclaimed funding
- 👥 Serve 2 million+ U.S. farmers
- AgFirst/SCSU Hackathon 2026 - Functional prototype built in 48 hours
- Target Audience: Young, Beginning, and Small (YBS) farmers
- Social Impact: Demonstrating how technology can strengthen rural communities through better funding access
Note: This project was created as a hackathon submission to demonstrate the concept and feasibility of an AI-powered funding discovery platform. It is not affiliated with or endorsed by the USDA.
This project is licensed under the MIT License - see the LICENSE file for details.
Built with ❤️ for farmers by the Cultura team.
- Team Lead:
- Dajanique Leysath
- Team Members:
- Zaria Henderson
- Dajie Weeks
- Jacob Alexander
- Elijah Mack
- Carlos Knight
- LinkedIn: Dajanique Leysath
- South Carolina State University - Hackathon host
- AgFirst Farm Credit - Hackathon sponsor
- AWS - Cloud infrastructure and AI/ML services
- USDA - For making federal spending data publicly available
- USAspending.gov - Database of federal programs
Made with 🌱 to help farmers grow