Apratim Chaudhuri

Why “quick automation” often breaks later

I’ve seen many “quick automation” scripts fail after a few weeks — not because of Python, but because edge cases were ignored.

Common misses:

Empty or partial inputs Unexpected formats Duplicate or inconsistent data

Before writing code, I...

Why I avoid over-engineering automation

Not every automation needs a framework, CI/CD, Fancy dashboards

Sometimes a single Python file with clear logic is enough.

Over-engineering increases Maintenance, Fragility and cost

The best automation is the one that keeps working quietly.

During testing, we needed specific VINs with multiple trim options.

Manually finding them was painful and slow.

I built a small Python script that:

Pulled VIN prefixes,Queried available trims,Filtered only VINs with 10+ trim options

It wasn’t a “product” — just a focused...

No API. No DB. No UI automation. Still delivered value. In one project, automation was blocked from all sides: No database access, No APIs, UI automation rejected due to security

Instead of forcing tools, I built supporting utilities:

Data extractors Log analyzers Input validators

...