Benchmarked AI

I'm Amo. I've spent 17 years in IT and datacenter infrastructure — first in the UK, now in Switzerland. I've been tinkering with AI tools since long before "AI tools" was a product category, and my background in Linux, hardware, and large-scale infrastructure gives me a different lens on these tools than most reviewers have.

This site exists because most AI tool reviews are written by people who haven't used the tools. They summarize feature lists, repeat marketing copy, and call it a review. I benchmark everything I write about — real workflows, real timing, real cost math. I approach every tool the way I'd evaluate a new piece of infrastructure: what does it actually do under load, what breaks first, and is the pricing model honest?

Every article on this site includes firsthand testing evidence. Every screenshot is from my own session. Every number is real. If I couldn't test it, I don't publish it.

What I test

AI app builders, automation platforms, presentation tools, writing assistants, analytics tools — anything a working professional might actually pay for. I focus on tools where the gap between the marketing and the reality is worth documenting.

How I work

I test with real use cases, not toy examples. I track time. I note what breaks. My datacenter background means I pay close attention to pricing models — credit systems, token traps, and upgrade walls are a recurring theme in this space and I call them out directly.

Affiliate links

Some articles include affiliate links. I only link to tools I've genuinely tested and would recommend. Commissions don't change my ratings — if a tool is mediocre, I say so. See the full affiliate disclosure.