From 27 years in the trenches to a Public Beta: Launching mAPI-ng

I’ve been working in IT since 1997. Over nearly three decades, and the last 8 years running my own boutique consultancy, Doolta. I’ve lived through every evolution of the ‘production incident.’

Whether it was Perl, Java, or now Go, the nightmare is always the same: an API starts failing, and you’re trapped staring at a wall of dashboards, guessing the root cause.

I got tired of the ‘Observability Tax.’ Setting up and understanding a full OTel + Prometheus + Grafana stack shouldn’t feel like a second full-time job.

So I built the tool I wish I had 20 years ago.

Today, I’m opening the public beta of mAPI-ng (monitored API, NextGen).

What is it?

mAPI-ng is an evidence-backed diagnosis engine for Go APIs. Instead of just showing graphs, it automates the heuristics I’ve developed over 27 years of debugging. It correlates RED metrics with Go runtime signals to rank the most likely causes (GC pressure, connection pool congestion, IO bottlenecks, etc.).

Why it’s different:

• Veteran Logic: It doesn’t just show data; it interprets it like a senior SRE would.
• Total Freedom: The entire stack is MIT licensed. I recently relicensed everything from BSL to MIT because I believe in convenience, not lock-in. You can self-host it with a simple make up.
• Zero-Config: Two imports, one middleware. That’s it.

The Mission:

The hosted version at mapi-ng.com is for those who want the power of ClickHouse-backed observability without the operational headache.

I’m looking for my first 100 beta users to stress-test the diagnosis engine. If you’re running Go in production and want to see if your API can finally diagnose itself, I’d love your feedback.

It’s forever free for small projects, no credit card required.

To my fellow ‘long-timers’ in the industry: what’s the one incident that changed how you think about monitoring?"