System Profile // About
AMIT
CHHABRA
Cloud Architect engineering systems that don't go down even when everything around them does. Based in Delhi NCR, working with teams worldwide who treat uptime as a feature, not an accident.
The Architect
I design AWS infrastructure for teams that cannot afford a bad night. That means fintech platforms processing live transactions, healthcare systems that need to be reachable at 3 AM, and e-commerce stacks that meet their biggest traffic spike on the worst possible day.
My work sits at the intersection of architecture and discipline multi-AZ failover, blue-green deployments, autoscaling that actually scales, and monitoring that tells you about a problem before your customers do. Zero downtime isn't a single feature; it's a hundred small decisions made correctly and consistently.
Outside of client work, I write about distributed systems and contribute to open-source tooling for infrastructure-as-code. I'm based in Delhi NCR and work with teams across time zones.
Deployment Log
Career changelog, most recent first
Independent Cloud Architecture consultant
Working with growth-stage teams to design fault-tolerant AWS infrastructure that stays up while everything else fails around it.
Formalized zero-downtime deployment practice
Built repeatable blue-green and canary release pipelines, cutting client deployment incidents to near zero.
Specialized in High Availability
Started designing multi-AZ and multi-region systems for clients who could not afford downtime - fintech, healthcare, e-commerce at scale.
Moved into AWS full-time
Rebuilt monolith infrastructure on EC2, RDS, and S3. Picked up the AWS Solutions Architect certification along the way.
Started as a backend engineer
Shipped first production services. Learned the hard way that "it works on my machine" is not a deployment strategy.
Architecture Tenets
Principles every system I build has to pass
Design for failure
Every component will eventually fail. The system should already know what to do when it does.
Redundancy over heroics
No architecture should depend on someone being awake at 3 AM to keep it alive.
Automate the boring
If a human has to repeat it, it belongs in a pipeline not in a runbook.
Measure everything
You cannot improve what you do not monitor. Observability is not optional.
Redundancy Systems
Toolchain, by failover layer
- EC2 Auto Scaling
- ECS / Fargate
- Lambda
- Route 53
- CloudFront
- ALB / NLB
- RDS Multi-AZ
- DynamoDB Global Tables
- S3
- Kubernetes
- Terraform
- CI/CD (GitHub Actions)