Toggl was fine. I paid the monthly fee, ran timers, pulled reports, stitched the data into QuickBooks invoices by hand. Fine is a word that does a lot of work.
Then I started running parts of my business with AI agents, and “fine” stopped being good enough. Toggl has an API. It has integrations. But it was not built for an AI to drive end to end, and the pieces I cared about – rounding rules, an invoice export tuned to QuickBooks, reporting around how I actually bill – were never showing up on the roadmap.
So I built my own.
This series walks the full arc: why I left Toggl, how the tracker is shaped so my AI assistant operates it the same way I do, and what running it day to day actually looks like.
Posts are listed below in the order they were published.
Posts in this series
Building Your Own AI-Ready Time Tracker
Why I replaced Toggl with a self-hosted, API-first time tracker that my AI assistant can operate end to end.
Apr 18, 2026The Time Tracker, In Practice
A visual walkthrough of the self-built time tracker I replaced Toggl with, from a running timer to an emailed QuickBooks invoice.
Apr 19, 2026Starting Timers from Inside Trello
A Chrome extension that adds a start/stop button inside Trello cards, so the time tracker meets work where it actually happens.
