ENGINEERING NOTES
Technical decisions
from shipped apps.
Notes on app architecture, native and cross-platform choices, AI features, release work, and controlled workflow agents.
Native iOS app or wrapper? A decision guide for product teams
Choose based on the workflows the iPhone app must support and the Apple frameworks it needs.
Read the noteOn-device AI for iOS and macOS: what should run locally?
Use privacy, latency, cost, resilience, model lifecycle, and hardware limits to choose a local, cloud, or hybrid architecture.
Read the noteHow to scope an AI workflow agent with permissions and evals
Define the task, permitted actions, test cases, and human review before choosing how much the system can do on its own.
Read the noteApp Store release checklist for an AI-generated prototype
A working screen still needs failure handling, subscriptions, privacy declarations, device testing, and a repeatable release process.
Read the noteFrom model demo to native macOS AI app
A Mac product needs model lifecycle management, memory guards, permissions, updates, licensing, and failure recovery.
Read the noteDISCUSS YOUR APP
Have a similar
engineering decision?
Send the current product, architecture, or workflow and the decision you need to make.
Discuss your app