1 Comment

User's avatar
Pawel Jozefiak's avatar

120 companies building agents, but the thing I keep wondering: who's building the management layer?

Running a personal AI agent taught me that the agent itself is maybe 30% of the work. The other 70% is observability - knowing what it did, what it decided, what it got wrong. I literally outgrew a spreadsheet tracking my agent's tasks and had to build a proper dashboard, then a native app.

The unglamorous truth: most agent failures aren't capability failures. They're visibility failures. You can't fix what you can't see running.

My infrastructure journey from spreadsheet to native app: https://thoughts.jock.pl/p/wiz-1-5-ai-agent-dashboard-native-app-2026

No posts

Ready for more?