One bounded decision
One workflow, one production question and a named executive sponsor keep the Mandate commercially and operationally precise.
AI systems architecture · Vienna & DACH
The pilot worked in the room. Production asked harder questions.
The demonstration lands well. Then the operations lead asks who owns the decision, who handles the exception and who can stop the workflow when the evidence is weak. The answer moves around the table.
The Architecture Mandate is a fixed-scope advisory engagement for one priority workflow and one production decision. It turns that uncertainty into an evidence-based executive decision and an inspectable operating architecture.
Decision signals
Five possible outcomes
The answer may be Proceed, Proceed with conditions, Redesign before pilot, Do not proceed or Evidence insufficient.
Advice before implementation
The Mandate defines the operating boundary, controls and evidence needed for a later decision; it does not promise a pilot or deployment.
Fee and engagement
A bounded mandate with explicit commercial terms
€22,000 excluding VATFixed fee for the standard scope. Engagement follows qualification, an approved proposal and a signed Statement of Work.
Request an Architecture Mandate proposal- Delivery window
- 15–20 business days after mobilisation conditions are met.
- Invoice schedule
- 40% on signature, 40% after the working session and draft architecture, and 20% before the final Executive Decision Pack.
- Payment term
- Invoices are payable within 14 calendar days.
The operating problem
A persuasive model can still sit inside an unowned production decision
The problem often appears in an ordinary meeting. A non-standard case arrives, the output looks plausible, and responsibility starts moving between operations, IT, management and the vendor.
The Architecture Mandate starts with one real workflow rather than a platform shortlist. We reconstruct the decision from request to outcome: what evidence is required, what the system is allowed to do, who can accept or reject its recommendation, and where work goes when data is absent, contradictory or outside the intended context. That reveals the difference between a useful capability and an operable system.
We then make accountability part of the architecture. A decision and authority matrix names the workflow owner, the subject-matter reviewer, approval and override rights, the exception destination and the person with stop authority. An evidence plan defines what must be recorded. A baseline describes the current flow before a pilot is allowed to claim improvement.
This work respects the EU AI Act’s risk-based structure; it does not label every AI workflow high-risk. Where a system is in fact classified as high-risk, the specific provisions on lifecycle risk management, record-keeping and effective human oversight become relevant within their legal scope.
System method
Move from a promising pilot to a defensible production decision
NIST AI RMF and TEVV are voluntary operating references, not certification or proof of EU conformity. The Mandate uses them to connect business consequence, human authority and technical evidence across four disciplined phases.
-
01
Mobilise around one decision
Confirm the executive sponsor, priority workflow, decision question, legal entity, primary country, evidence location and decision participants before the delivery clock starts.
-
02
Map workflow, authority and dependencies
Trace real cases, identify material systems and hand-offs, and name approval, override, suspension, release and escalation authority.
-
03
Design controls, evidence and target architecture
Set human review, exception and stop paths; map material risks and the evidence each control must produce; compare architecture options without pretending assumptions are facts.
-
04
Hold the executive decision workshop
Present the evidence quality, recommended architecture, rejected alternatives and unresolved conditions, then record exactly one of the five permitted decision outcomes.
What the work produces
Five deliverables that make the production decision inspectable
The Mandate ends with working material, explicit limits and one recorded decision. Acceptance confirms delivery completeness; it does not certify the system or transfer management accountability.
Executive Decision Pack
The production decision, business and investment logic, evidence limits, build-buy-integrate recommendation, conditions, owners and next authorised step.
Workflow and Authority Model
The current workflow, actors, hand-offs, exceptions and decision rights for approval, override, suspension, release and escalation.
System, Data and Dependency Map
In-scope systems, data origins and destinations, interfaces, ownership boundaries, vendor dependencies and unverified single points of failure.
Risk, Control and Evidence Map
Material risks, preventive and detective controls, control owners, triggers, required evidence, escalation and residual uncertainty.
Target Operating Architecture and Decision Record
The target boundary, roles, options considered, decision rationale, conditions and acceptance gates for any separately authorised next phase.
Illustrative operating scenario
Illustrative scenario — not a customer caseA good recommendation with nobody authorised to own it
Illustrative scenario — not a customer case
A Vienna-based B2B team pilots an AI assistant to prepare complex proposal approvals. The recommendations look useful in the steering meeting. The first unusual case, however, contains stale contract data. Sales assumes Legal will review it; Legal expects the system to escalate; IT can confirm only that the service ran. The Mandate does not add another approval button. It names the decision owner, defines the actions the assistant may take, sets the minimum evidence for review and gives the exception a destination. The executive team can then decide whether to proceed, impose conditions, redesign, stop or request missing evidence.
Purpose: explain the method. No operating outcome is claimed. The scenario demonstrates the method only and claims no realised operating result.
Execution roadmap
A bounded Mandate from mobilisation to executive decision
-
Mobilisation
Confirm scope, access and evidence readiness
The delivery window starts only after the workflow, sponsor, participants, evidence location, access conditions and working dates are confirmed.
-
Evidence and architecture
Build the operating view
Interviews, evidence review and one working session expose the current flow, authority, dependencies, risks, controls and viable architecture options.
-
Executive decision
Choose one documented outcome
The executive workshop tests the recommendation and records the outcome, conditions, accountable owners, evidence gaps and next authorised step.
Questions worth resolving
Questions worth resolving
What is the Architecture Mandate?
It is a fixed-scope advisory engagement for one priority workflow and one production decision. It creates an evidence-based executive decision, an operating architecture and explicit conditions for any later phase.
How is this different from an AI Systems Review?
A general review can collect a broad set of technical and organisational questions. The Architecture Mandate is narrower: for one workflow it ends with exactly one documented production decision, defined deliverables and explicit boundaries.
What fits within the standard scope?
The standard scope covers up to five material systems, up to eight stakeholder interviews, one working session, one executive decision workshop, up to 25 representative artefacts and two consolidated review rounds. Delivery is planned for fifteen to twenty business days after readiness conditions are met.
What decision does the Mandate produce?
Exactly one state is recorded: Proceed, Proceed with conditions, Redesign before pilot, Do not proceed or Evidence insufficient. Missing or conflicting evidence is recorded rather than converted into an assumption.
What must the client provide?
The client appoints an executive sponsor and coordinator, confirms the workflow and decision question, provides lawful and appropriately minimised evidence, makes participants available and returns consolidated factual feedback.
What is outside the Mandate?
It excludes model development, implementation, integration, migration, deployment, managed operations and specialist legal or security assessments. The Mandate does not certify legal, security, privacy, regulatory, financial, technical or production outcomes.
Primary references
References inform the operating context; they do not imply endorsement.
- European Commission — EU AI Act risk-based framework
- EU AI Act, Article 4 — AI literacy
- EU AI Act, Article 9 — risk management for high-risk AI
- EU AI Act, Article 12 — record-keeping for high-risk AI
- EU AI Act, Article 14 — human oversight for high-risk AI
- NIST AI Risk Management Framework 1.0
- NIST AI Test, Evaluation, Validation and Verification
Request an Architecture Mandate proposal
Bring one AI production decision that cannot responsibly stay vague.
The first conversation tests whether the Architecture Mandate fits the decision, evidence and executive ownership in front of you. It is a qualification conversation, not a free audit and not a promise that a pilot will proceed.