Concepts

For the Curious

You don't need this section to use fai. Everything works without understanding why. But if you're the kind of person who wants to know what's underneath - what ideas shaped the design, what principles constrain every decision - this is where they live.

Your Workbench

The central metaphor. A master carpenter's workbench is personal - tools arranged their way, jigs built for their projects, techniques perfected over years. When they change workshops, the workbench goes with them.

Your AI workbench works the same way. What travels: your patterns, your preferences, your professional identity. What stays: the work you built.

Context

The word we use instead of "memory" - and the distinction matters. Memory is opaque, locked inside a model, owned by the provider. Context is yours. You can see it, change it, export it, delete it. It's an artifact, not a weight.

Progressive Growth

Why fai reveals capabilities layer by layer instead of dumping everything on you at once. Your framework doesn't come from a catalog - it emerges from your work, curated by you, at your pace.

Architecture Boundary

The operating principle that constrains every design decision at fai: does the user see architecture or experience? If architecture leaks through, we fix the architecture - not the docs.

Resilience

Why we say "flow protection" instead of "uptime." When you're deep in creative work, any interruption costs 23 minutes of recovery. fai is designed to make sure your AI is always there - switching providers before limits hit, preserving context across the handoff.

These Ideas in Practice

The concepts here aren't theoretical - they show up in every page of the docs. If you want to see them in action across different domains, explore the Examples.