Gendering AI — four passes
Put a real system in front of you and ask who was counted, who was harmed, who decided and who pays. Works fully offline once opened.
Systems that keep working when the connection does not.
I build learning, records and commerce systems for contexts where a stable connection is the exception. That constraint changes the design: the work has to live on the device, resume without a signal, and only reach the server when the person decides it should.
Every open journey below is free to use, needs no account, and holds your answers locally rather than on a server.
Open journeys
Each one is a working session, not an article. You leave with something written down.
Field recon for researchers
Collect notes in a place with no signal; the record stays on the handset and files itself when you are back. Launching next.
First move for a home-based business
Early recon on what to build, what to buy and who can mentor you — the same journey engine, pointed at a different question. Launching next.
Why offline is the interesting constraint
A dropped connection is an inconvenience for someone on fibre. For a researcher in the field or a parent building a business between other work, it is the difference between finishing and not finishing. Designing for the drop is not a fallback feature — it is the test of whether the work was ever really possible on the user’s terms.