Choosing Your Tech Stack: A Founder's Decision Framework

One of the most common conversations I have with founders is about tech stack selection. They come with a list of technologies they have heard about (Rust, Elixir, some new database) and want my opinion on which to use. My answer is almost always the same: use the boring option.

The boring technology thesis

Boring technology, meaning well-established, widely-used, extensively-documented tools, has a massive advantage: it lets you focus on your product instead of fighting your infrastructure. React, PostgreSQL, Node.js, Python. These are boring. They are also reliable, well-understood, and have massive ecosystems.

Every novel technology in your stack is a place where you will spend time debugging documentation gaps, finding workarounds for missing features, and explaining to future developers how things work.

The decision framework

For each technology choice, ask three questions. First: can we hire for this? If you are building a team, the pool of developers who know React is 50 times larger than the pool who know Svelte. Second: will this technology still be supported in 3 years? Third: does our team already know it?

If the answer to all three is yes, use it. If the answer to any is no, you need a very compelling reason to proceed.

When to use something new

There are legitimate reasons to choose newer technology. If your product has a genuine technical requirement that boring options cannot meet (real-time at massive scale, specific AI capabilities, hardware integration), then the newer option is justified.

The key is being honest about whether the requirement is real or aspirational. Building for 'future scale' with complex infrastructure when you have 100 users is a common trap.