If I had to start over today, I'd keep it painfully simple:
Build stuff but also focus on marketing
Most developers (myself included) fall into the trap of tweaking features, polishing deploy scripts, and 'optimizing infra' way too early.
But in the early days, the real work is figuring out what people actually care about, and how fast you can get paid to solve it.
Here's what I wish someone told me earlier:
Launch before you're ready.
Start charging as soon as it works.
One Droplet, managed Postgres, maybe Redis, don't over-engineer.
Use boring tech you can support solo.
Spend more time on writing and talking than coding.
Anyway, if you're building something in 2025, what are you doing differently?
Let's trade notes.
(P.S. I recently redesigned DevOps Daily so I have to start doing marketing 🤪)