Technology
The adaptive architecture behind Dryvn.
Dryvn’s development approach starts with a question: which parts of an operation are shared, and which parts need to adapt? DAAOA brings that distinction into the design of an AI operating system for business.
This page explains the approach and where it stands today.
- 01Business context. Customers, work, documents and history the operation already has.
- 02Configured capabilities. The functions, terminology and workflows enabled for this organization.
- 03Connected systems. Software the business already runs, where the implementation supports it.
- 04Operator review. Permissions, approvals and responsibility for exceptions.
What is DAAOA?
DAAOA, Dryvn Adaptive Archetype Operating Architecture, is Dryvn’s approach to configuring an AI operating system around the way a business works. An operating archetype describes a recurring pattern of work, such as managing a job or project. Shared capabilities can then be configured with the terminology, workflows, rules and permissions appropriate to the organization.
Dryvn’s job and project operating model is its most mature starting point. Broader archetype support and further adaptation are development areas; availability depends on the capabilities and workflows verified for each implementation.
What is an operating archetype?
An operating archetype is a recurring pattern in how a business organizes and completes work. Industry describes the market a company serves; an archetype describes how the work moves. Different industries can share a job or project pattern while using different terminology, rules and specialist capabilities.
Text version of this diagram
- Request (start) → go to Scope
- Scope (step) → go to Estimate
- Estimate (step) → go to Approved?
- Approved? (decision) → yes: go to Scheduled work; not yet: go to Follow up or expire
- Scheduled work (step) → go to Documentation
- Documentation (step) → go to Billing
- Billing (end)
- Follow up or expire (waiting state) → revised: go to Estimate
Illustrative pattern. A renovation contractor, a sign shop and a landscaping company can all run this shape with different words for each stage. Which stages are enabled is set per implementation.
What stays shared, and what changes?
| Shared foundation | Configured experience |
|---|---|
| Business records and relationships | Terminology and relevant fields |
| Available operating capabilities | Enabled capabilities and workflow choices |
| Mechanisms for permissions and review | Organization-specific permissions and approval requirements |
| Connections supported by the implementation | Selected systems and agreed ownership of records |
This is an architecture principle. It does not mean every archetype is implemented to the same depth, and it is not an enforcement guarantee; which controls apply is part of each implementation.
What does adaptive mean?
Adaptation can include approved terminology, preferences, workflow configuration and operating rules. It does not mean that the system silently rewrites business policy. Dryvn is being developed to use approved operating context more effectively over time; the exact scope depends on the capability and implementation.
Where do people remain in control?
An implementation needs clear authority for actions, named responsibility for exceptions and evidence that required steps were completed. Review which controls are currently available and how they apply to the workflow before relying on unattended execution.
What is available, and what is being developed?
- Most mature starting point
- The job and project operating model: request, scope, estimate, scheduled work, documentation and billing, with follow-up and approvals around it.
- What the current product holds and does
- Development areas
- Broader archetype support, further adaptation from approved operating context, and additional connections. Described as direction until verified for an implementation.
- Integration status, entry by entry
No release dates, customer counts or delivery promises are published here.
Understand the underlying concepts.
Educational explainers from the Dryvn library.
- What is an AI system, a workflow and a loop?how a system, a workflow and a loop differ
- How AI uses your documents: search, context and retrievalhow AI uses documents and records
- Knowledge graphs and workflow graphs, explained for business ownersknowledge graphs and workflow graphs, explained
- How multi-model AI systems work, and when one model is enoughwhen several models help and when one is enough
- What should AI be allowed to do without asking?how to set approval boundaries
See the product the architecture describes, or bring one workflow and discuss the fit.