Engineer: "Hey AI, our payment service is getting slow. Can you scale it up by 50%?"
AI SRE: "I'll scale the payment service for you. Let me check the current status first..."
Current: 4 instances (t3.large)
CPU: 78% average
Memory: 62% average
Request rate: 1,200 req/s
Scaling to: 6 instances
Estimated completion: 2 minutes
"Scaling initiated. I'm also noticing high CPU usage. Would you like me to upgrade to t3.xlarge instances for better performance?"
Engineer: "Yes, and also set up auto-scaling for the future"
AI SRE: "Done! Here's what I've configured:"
✓ Upgraded to t3.xlarge (6 instances)
✓ Auto-scaling enabled:
- Scale up: CPU > 70% for 5 min
- Scale down: CPU < 30% for 10 min
- Min: 4, Max: 20 instances
✓ CloudWatch alarms configured
✓ Estimated monthly savings: $1,200
Terraform code generated and applied.