An AI-powered serverless agent that answers football (soccer) questions using AWS Lambda and Claude AI. Built as a learning project to explore serverless architectures and AI integration.
curl -X POST https://p3ex1w73fg.execute-api.us-east-1.amazonaws.com/prod/ask \
-H "Content-Type: application/json" \
-d '{"query": "When does Real Madrid play next?"}'Other queries you can try:
"What's the score in the Arsenal match?""Who is leading La Liga?""Tell me about Manchester City's performance"
A serverless API that understands football-related questions and provides intelligent, conversational responses. Built with AWS CDK for infrastructure as code and designed to be cost-effective (~$6/month for 1000 requests).
User Query → API Gateway → Lambda Function → Claude AI
↓
Football Data
Processing
- AWS API Gateway: RESTful API endpoint with rate limiting and CORS support
- AWS Lambda: Serverless compute running Node.js 18.x
- Anthropic Claude API: Advanced language model for natural response generation
- AWS CDK: Infrastructure as Code for reproducible deployments
- Natural Language Queries: Ask questions in plain English about matches, scores, and teams
- Intent Detection: Smart recognition of query types (next matches, live scores, standings)
- Context-Aware Responses: Combines structured data with AI-generated insights
- Error Handling: Graceful fallbacks with football-themed error messages
- Production-Ready: Rate limiting, logging, and monitoring built-in
"When does Real Madrid play next?"
"What's the current score in the Arsenal match?"
"Show me the La Liga standings"
"How has Manchester City been performing?"
- Infrastructure: AWS CDK (TypeScript)
- Runtime: Node.js 18.x
- AI Model: Claude 3.5 Sonnet
- Cloud Services: Lambda, API Gateway, CloudWatch, Secrets Manager
- AWS Account with appropriate permissions
- Node.js 18.x or higher
- AWS CLI configured with credentials
- AWS CDK CLI (
npm install -g aws-cdk) - Anthropic API key
- Clone the repository:
git clone https://github.com/vnponce/football-intelligence-agent.git
cd football-intelligence-agent- Install dependencies:
npm install- Set your Anthropic API key:
export ANTHROPIC_API_KEY="your-api-key-here"- Deploy to AWS:
cdk deploy --profile your-aws-profileAfter deployment, test your agent:
curl -X POST https://your-api-url/ask \
-H "Content-Type: application/json" \
-d '{"query": "When does Barcelona play next?"}'Estimated monthly costs for moderate usage (1,000 requests):
- API Gateway: ~$0.001
- Lambda: $0 (within free tier)
- Claude API: ~$5-6
- Total: ~$6/month
Starting with hardcoded data allowed focus on:
- Core AI integration and response quality
- Infrastructure patterns and best practices
- Predictable demos and testing
- Faster iteration without external dependencies
- Superior context understanding for sports queries
- Natural, engaging response generation
- Reliable API with good documentation
- Cost-effective for demonstration purposes
- API keys stored in environment variables (Secrets Manager ready)
- CORS configured for web access
- Rate limiting to prevent abuse
- Request validation and sanitization
- Integration with live football data APIs
- Multi-language support
- Team-specific notification subscriptions
- Historical match analysis
- Prediction capabilities based on form data
- Voice interaction support
This project demonstrates:
- Building production-ready serverless applications
- Integrating AI services into practical applications
- Infrastructure as Code best practices
- Cost-conscious cloud architecture design
- Error handling and user experience in AI applications
Feel free to open issues or submit pull requests. Areas for contribution:
- Additional team/league data
- New query patterns and intents
- Performance optimizations
- Test coverage improvements
MIT License - feel free to use this code for your own projects!
Abel V.
Senior Full-Stack Engineer specializing in AWS and Applied AI
Built with passion for football and technology ⚽❤️