But in real projects, the hardest questions are still unsolved:
- How do you structure a project for a team?
- How do you build CI/CD with real rollback?
- What does a safe migration strategy look like?
- How do you set up observability before things break?
- How do you actually debug production?
I spent a long time figuring this out across different projects,
and couldn’t find a single place that connects all of it together.
So I built a simple Todo API as if it were a real production system.
The focus is everything around the code:
CI/CD, Terraform on AWS, observability, release automation, rollback.
Not a boilerplate. A reference with reasoning behind every decision.
Would love feedback from people who've dealt with this in production.
AI made writing backend code much easier.
But in real projects, the hardest questions are still unsolved:
- How do you structure a project for a team? - How do you build CI/CD with real rollback? - What does a safe migration strategy look like? - How do you set up observability before things break? - How do you actually debug production?
I spent a long time figuring this out across different projects, and couldn’t find a single place that connects all of it together.
So I built a simple Todo API as if it were a real production system.
The focus is everything around the code: CI/CD, Terraform on AWS, observability, release automation, rollback.
Not a boilerplate. A reference with reasoning behind every decision.
Would love feedback from people who've dealt with this in production.