Building Ether solo: scope discipline over feature count

Ether started as a simple question: could I own an entire product end-to-end — design, engineering, and the decisions in between — without a team to lean on? The constraint that shaped every other decision was time: this was a solo bachelor's thesis build, not an open-ended side project.

The biggest call was persistence. A cloud-synced backend would have made the app feel more 'production,' but it would have also meant building and maintaining auth, sync conflict resolution, and infrastructure I didn't have time to do well. I chose CoreData instead — local-first, simple, and it let me spend the time budget on the parts that actually differentiated the app: the guided yoga/meditation flows, the auto-generated daily planning, and the meal-planning module.

That trade-off is easy to defend in hindsight: a finished, coherent local-first app beats a half-wired cloud app every time, especially against a fixed deadline. But it's a real trade-off, not a free lunch — Ether doesn't sync across devices today, and that's the honest cost of the choice.

Working through the whole stack solo — UI, state, persistence, and the product thinking connecting them — changed how I scope data projects too. I'm now much quicker to ask 'what's the smallest complete version of this' before reaching for the more sophisticated architecture.

If I extended Ether today, CloudKit sync and a watchOS companion are the obvious next steps — both are additive without requiring a rewrite of what's already there.