Sketches / teaching

Local, hosted, and managed AI architecture

A teaching sketch for separating client, API, model, and inference boundaries when a full-stack product uses local, hosted, or managed AI services.

mentoringfull-stackai-systems
full-stack-ai-architecture.excalidraw read only
Loading sketch…

Pan and zoom with Excalidraw gestures

Context

The drawing carries the visual model.

This sketch was drawn during a full-stack architecture instruction session. It separates client, API, model, database, and inference choices so a team can name which pieces are local, self-hosted, or managed by a provider.

Provenance

Used in

Atlas context

Nearby in the Atlas.

See how this role connects to the projects, case studies, notes, and ideas around it.

Explore the full Atlas →
overview · 72%