Advice to New Devs
Yesterday I met with my junior dev to review some new features we’re adding to a new module in our ERP migration app. As we went through the requirements, he kept asking:
“What if… we’re not covering all the possible cases?”
I told him:
“Let’s look at this SQL result. How many rows do you see that are actually ‘what-if’ scenarios? Roughly 10%, right?”
That doesn’t mean edge cases don’t matter — it means you shouldn’t START by trying to solve them all at once.
Here’s the approach I recommended:
  1. Build the general case first
  2. Test the heck out of it
  3. Identify edge cases (make an explicit list)
  4. Implement edge cases one by one
Why this works:
If you implement everything at once and something breaks, you don’t know where the logic went wrong. And the worst bugs aren’t syntax errors — they’re incorrect logic.
AI is great at fixing broken code. But if you inserted bad logic, that one is on you.
For simple features, this might feel like overkill. But for complex systems (ERP, migrations, multi-role apps, etc.), this incremental approach is how you stay sane and ship reliably.
Hope it helps as you keep growing as a dev.
5
1 comment
Andres Dominicci
4
Advice to New Devs
Builder’s Console Log 🛠️
skool.com/ai-for-your-business
We build things because it’s fun.
console.log ("democratize developing");
Leaderboard (30-day)
Powered by