Sean's Dev Blog

On Choosing the Best Tech Stack

November 17, 2020

So you have this great idea. You’re going to scale it up, rake in some sweet investor dough, and buy a Lambo so all your friends can stay up late thinking about how cool you are. You have a lot of high ambitions.

The problem is you don’t have any code written, and you need to know how you are going to put this thing together. As part of the initial design phase, your first technical decision is inevitably which stack to use.

LAMP, ME[A/R/V]N, C#/.NET, serverless; the options can be overwhelming and can take months to decide.

It’s better to just pick a stack and run with it. Pick literally anything. If you have to rewrite it in another stack later, that’s fine. Pick something and build the minimum viable product as quickly as possible.

You can always optimize after you ship. The important thing is to deploy it and get users, ASAP. Collect statistics and run experiments. Take note of what works and what doesn’t. Refactor and rewrite when necessary.

Version 2 should, in principle, be better than version one. And if you can’t make it past version one, not even the best stack in the galaxy can save you.

Musings on coding, UX/UI, and hacker culture

© 2020 Sean Boisselle