Skip to content

Blog

Field notes from putting AI-built apps online.

What we measured, what we got wrong and what we changed. Each post carries the numbers we actually have, and links the guide or the docs that go deeper.

What goes here

Three kinds of posts, and nothing else.

  • A number from our own records. A funnel that did not convert, a scan over apps people deployed, a limit that bit someone — read from Tofu's own database rather than estimated.
  • A behaviour that confuses people. The order a database and an app arrive in, a switch nobody can find, a sentence on a page that reads one way and works another.
  • A change we made, and why. What the old behaviour was, what the new one is, and what we have not tested yet.

Step-by-step answers live under Guides, and the side-by-side comparisons under Compare. The docs explain each step in full.