Unclassified

Layer model

IntegrateWise is not one undifferentiated “AI product.” It is a layered system with a deliberately narrow user-facing expression.

LayerWhat it isWhat the user experiences
ExperienceThe interaction promise“My work already knows its context; AI prepared the hard parts; I decide.”
ProductThe role-composed IntegrateWise User WorkbenchA calm work surface for the person’s current role and work.
PlatformContinuity infrastructure for connected human and AI workMostly invisible: it preserves context, evidence, policy, and handoffs.
Access / distributionIW-Continuity-Bridge and approved entry pointsThe same bounded continuity where a user or approved external surface needs it.

The composition axis

The product is composed around human work context, not around an entity record, a conversation, or a generic AI surface:

role + department + active focus + canonical work state + authority + available capabilities = workbench

An entity can be part of the active focus. It is never the product’s organizing abstraction on its own.

This applies to every application category: Leads, Accounts, Marketing, Product, Finance, Service, Education, Personal, and future role domains. A category may have different work objects and capability slots; it must use the same continuity contract.

Ownership boundary

The Workbench owns presentation and deliberate user interaction. The platform owns durable truth, policy, orchestration, and governed execution. IW-Continuity-Bridge owns bounded access to platform capabilities. The Twin is a contextual cognitive participant in the product; it owns none of those system authorities.

Decision rule

When a proposal mixes layers, classify it before building it:

  • If it preserves or reconciles work across people, tools, and agents, it belongs to the platform.
  • If it changes what a person sees and decides while doing a role’s work, it belongs to the product.
  • If it packages platform capabilities for an approved external entry point, it belongs to access/distribution.
  • If it creates a persistent conversational destination or substitutes AI for the decision-maker, it violates the intended experience unless the user explicitly enters a deeper Ask flow.