Building Klau in the Open, Week 10: The Lindy Effect

Week 10 of building Klau in the open.

I learned about the Lindy Effect this week. The longer something survives, the longer it is likely to keep surviving. Ten weeks in means I probably have ten more ahead. If that happens, twenty after that. The thing ages in reverse.

This week felt different from the ones before it. For weeks I had worked every free hour, changed direction many times, and second-guessed everything. This week was calm. I haven't felt that with Klau before.

Two paying customers now.

With real users on the system, real issues came up: new edge cases and things that aren't quite right. I spent a big part of this week finding bugs and fixing them. That may not sound exciting, but it is for me. A bug report from a user means they are using the software and care enough to tell you what's wrong. That feedback loop keeps me going. Every fix makes the system stronger and makes the work more real.

I also solved a problem I had been thinking about for weeks. I refined the dispatch optimization algorithm so it works for large fleets of hundreds of trucks. The earlier version would have worked, but its compute cost grew badly with fleet size. This one doesn't have that problem. I can build on it now and trust the foundation.

Ten weeks is still ten weeks. There is a lot more work before this pays for itself. But this week made me think Klau has a real chance.

Originally posted on LinkedIn on March 13, 2026.