An AWS-powered intelligent business analytics platform that provides real-time insights, AI-powered chat, and predictive simulations for enterprise operations.

Live Demo: http://insightsgridai-results-1761385095.s3-website-us-east-1.amazonaws.com/

• AWS CLI configured with appropriate permissions
• Node.js 18+ and npm installed
• AWS CDK installed: npm install -g aws-cdk

# Clone and navigate to project
cd /Users/vedantkandge/Documents/Hackathons/ABC

# Deploy infrastructure and frontend
./deploy.sh

The script will:

1. Deploy AWS infrastructure using CDK
2. Upload sample data to S3
3. Build and deploy React frontend
4. Display your website URL

• AI Chat Tab: General business intelligence queries using Airia pipeline
• Chat with Data Widget: Data-specific queries with contextual insights
• Natural language interface for complex business questions
• Real-time responses with intelligent caching

• Fuel Price Impact Analysis: Analyze cost effects of fuel price changes
• Demand Forecasting: Predict revenue and capacity requirements
• Warehouse Expansion ROI: Investment analysis for new facilities
• Route Optimization: Cost and time efficiency improvements

• Interactive data visualizations
• Key performance indicators (KPIs)
• Regional performance breakdowns
• Trend analysis and forecasting
• Custom metrics and filters

• Financial metrics (revenue, costs, profit margins)
• Logistics performance (shipments, delivery times, fuel costs)
• Sales analytics (products, regions, customer segments)
• AI-generated insights and recommendations

• S3: Data lake and static website hosting
• Lambda: Serverless compute (Python 3.11)
• API Gateway: HTTP API v2 & REST API
• DynamoDB: Session management and caching
• Athena: Serverless SQL queries
• Glue: Data catalog and ETL
• CloudWatch: Logging and monitoring

• Frontend: React 18, TypeScript, TailwindCSS, Recharts
• Backend: Python 3.11, AWS Lambda
• AI/ML: Airia AI platform with custom pipelines
• Infrastructure: AWS CDK (TypeScript)

Base URL: https://aij9r8sc8i.execute-api.us-east-1.amazonaws.com

• POST /chat - AI Chat interface (Airia pipeline: 0e2a6599)
• POST /chat-with-data - Data-aware chat (Airia pipeline: 70ef9a9d)

Base URL: https://h3qy1xq5kh.execute-api.us-east-1.amazonaws.com/prod/api

• GET /data?type=overview - Business overview metrics
• GET /data?type=trends - Trend data for charts
• GET /data?type=costs - Cost analysis by route
• POST /simulate - Run business simulations

The platform includes enhanced sample datasets:

• finance.csv: Revenue, costs, profit by date, region, department
• logistics.csv: Shipment data with carriers, routes, costs, fuel prices
• sales.csv: Product sales with categories, channels, forecasts
• customers.csv: Customer segments and demographics
• inventory.csv: Stock levels and warehouse data

Both chat features use Airia with specialized pipelines:

AI Chat Tab:

• Pipeline ID: 0e2a6599-9c5f-40ac-9a39-456a53b7d935
• Purpose: General business questions and analysis
• Timeout: 90 seconds

Chat with Data Widget:

• Pipeline ID: 70ef9a9d-5eb5-44e8-873b-e8060f024791
• Purpose: Data-specific queries and insights
• Timeout: 90 seconds

API Credentials: Configured in Lambda environment variables via CDK

Lambda functions automatically configured with:

• AIRIA_API_KEY: Airia authentication key
• AIRIA_USER_ID: User identification for pipeline execution
• ATHENA_DATABASE: Data catalog database name
• ATHENA_WORKGROUP: Query execution workgroup

cd frontend
npm install
npm start

The development server will proxy API requests to your deployed endpoints.

cd infrastructure
npm install
npm run build
cdk diff    # Preview changes
cdk deploy  # Deploy changes

# Test all API endpoints
./test-endpoints.sh

# Test chat features specifically
./test-both-airia.sh

# Comprehensive API testing
./comprehensive-test.sh

# Watch Lambda logs
aws logs tail /aws/lambda/insights-grid-bedrock --follow
aws logs tail /aws/lambda/insights-grid-airia-chat --follow

# Check API Gateway metrics
aws cloudwatch get-metric-statistics \
  --namespace AWS/ApiGateway \
  --metric-name Count \
  --dimensions Name=ApiId,Value=aij9r8sc8i \
  --start-time 2025-10-26T00:00:00Z \
  --end-time 2025-10-26T23:59:59Z \
  --period 3600 \
  --statistics Sum

• Answers in seconds, not weeks
• Real-time data processing
• Instant what-if scenario analysis
• Automated insight generation

• Predicts and prevents costly decisions
• Identifies optimization opportunities
• Reduces manual analysis time by 70%+
• Pay-per-use serverless architecture

• Handles enterprise-scale data
• Auto-scaling based on demand
• No infrastructure management
• Global availability

Monthly operational costs for mid-size company:

• Lambda: $50-100
• S3: $20-50
• Athena: $30-80
• DynamoDB: $20-40
• Airia AI: $100-300
• API Gateway: $10-30

Total: ~$230-600/month with 10x+ ROI potential

• IAM roles with least privilege access
• S3 bucket encryption at rest
• API Gateway authentication ready
• CORS configured for frontend origin
• Environment variables for secrets
• CloudWatch audit logging

Issue: Requests timeout after 29-30 seconds
Cause: API Gateway has 29s (REST API) / 30s (HTTP API) hard limits
Solutions:

• Simplify query complexity
• Check Airia pipeline performance
• Review CloudWatch logs for Airia 500 errors

Issue: Endpoints return 404 Not Found
Solution:

cd infrastructure
cdk deploy --force

Issue: Dashboard shows loading indefinitely
Check:

1. Browser console for API errors
2. API Gateway endpoints are deployed
3. Lambda functions are active
4. S3 data files are uploaded

Issue: "Internal Server Error" from Airia
Solutions:

1. Check Airia pipeline status at https://api.airia.ai/
2. Verify pipeline IDs in Lambda environment variables
3. Check Airia API key hasn't expired
4. Review CloudWatch logs for detailed errors

Issue: CDK deploy errors
Common Fixes:

# Clear CDK cache
cd infrastructure
rm -rf cdk.out/
npm run build
cdk bootstrap  # If first time
cdk deploy

• DEPLOYMENT.md: Detailed deployment guide and configuration
• ARCHITECTURE.md: Deep dive into system architecture and data flows
• TROUBLESHOOTING.md: Comprehensive problem-solving guide

To destroy all AWS resources:

cd infrastructure
cdk destroy --all

⚠️ Warning: This will permanently delete:

• All Lambda functions
• API Gateway endpoints
• S3 buckets and data
• Athena workgroups
• CloudWatch logs

• Factory performance simulations
• Supply chain optimization
• Capacity planning
• Cost impact analysis

• Inventory management
• Sales forecasting
• Regional performance analysis
• Customer behavior insights

• Route optimization
• Fuel cost management
• Delivery performance tracking
• Warehouse expansion planning

• Revenue forecasting
• Cost optimization
• Budget impact analysis
• Profitability modeling

• Real-time IoT data integration
• Advanced ML models for predictions
• Multi-tenant support with Cognito
• Mobile application
• WebSocket real-time updates
• Advanced visualization features
• Custom data source connectors
• API rate limiting and caching
• CloudFront CDN distribution
• Custom domain with HTTPS

This project is for educational and demonstration purposes.

This is a hackathon project. For questions or improvements, please open an issue or submit a pull request.

For issues or questions:

1. Check CloudWatch logs for errors
2. Review TROUBLESHOOTING.md
3. Verify AWS permissions
4. Check deployment script output
5. Review API endpoint status

Built with AWS Serverless Stack
Powered by Airia AI • React • AWS CDK
Last Updated: October 2025