How we work

One workflow. Four controlled stages. A result your team can operate.

We map the current work, build inside the systems already in use, stay through real operation, and measure against an accepted baseline. Reusable capability modules make the work faster; engagement-specific decisions keep it honest.

Map one workflow See the capability portfolio

What you get

Three things remain with your team

The map and accepted baseline

The current workflow, systems, exceptions, owners, decision boundaries, and starting measures in one portable artifact.

The operating workflow

A bounded implementation fitted to your systems, with model, deterministic, and human responsibilities separated explicitly.

The versioned evidence record

Acceptance criteria, operator decisions, exceptions, and measured observations retained as the baseline for the next decision. Later observations can update or invalidate that baseline.

Engagement timeline

Map, build, operate, measure — with named decision gates

A six-week implementation target is used when the workflow and dependencies support it. The Opportunity Map comes first, and the schedule is confirmed against the actual systems, owners, and authority boundaries.

Stage 1 · Map

Establish the workflow and baseline

Trace the current steps, systems, exceptions, owners, measures, and decision boundaries. Output: the Workflow Opportunity Map and a go, reshape, or stop decision.

Stage 2 · Build

Implement inside the existing stack

Configure the required capability modules, integrations, human gates, failure handling, and acceptance checks. VeilEngine is added only when the workflow requires its evidence boundary.

Stages 3–4 · Operate + Measure

Run with operators and compare to baseline

Handle real exceptions, tune the workflow, preserve human authority, and record what changed. The accepted evidence baseline is versioned for the next decision; later observations can update or invalidate it.

Where an engagement begins

One front door, two decisions that can follow

Every engagement starts by mapping one workflow. The map stands on its own; implementation and ongoing operation proceed only when the evidence supports the next commitment.

That front door is the same for established operations-heavy teams and founder-led businesses. Workflow volume, ownership, access, and measurable friction determine fit before company size does.

Entry · Typically 2–4 weeks

Workflow Opportunity Map

One current-state map, measured baseline, automation and control design, implementation brief, and a direct go, reshape, or stop recommendation. You keep every deliverable.

Map one workflow
Decision 2 · Bounded implementation

Build the selected workflow

Implement the approved design inside the existing stack with explicit human authority, acceptance criteria, failure handling, and a named operating owner.

See the delivery stages
Decision 3 · Operate + measure

Run, tune, and extend

Stay with the workflow through real use, resolve exceptions, measure against the accepted baseline, and decide whether the next workflow merits investment.

See the production field note

Durations are typical scopes, not fixed quotes. Commercial terms are set case by case against workflow complexity, system access, data boundaries, decision authority, and operating support.

Adjacent capabilities

Specialized surfaces in the portfolio

Three supporting capabilities extend the professional-service engagement for specific workflow classes. They do not replace the Workflow Opportunity Map or create separate public entry offers.

Document AI

Document Intelligence Layer

Layout-aware parsing, PII-safe ingestion, privacy-preserving retrieval-augmented synthesis. For real estate, legal, and grant-preparation workflows where the raw document itself carries the sensitivity.

Diagnostic surface

Workflow Opportunity Map

A fixed-scope diagnostic that maps one workflow, its systems, exceptions, owners, baseline, and decision boundaries before an implementation is funded.

Map one workflow
Public diagnostic

AI Governance Readiness

A public diagnostic surface measuring AI governance posture against a recognized AI-governance framework. Captures consent, scores readiness, routes to the appropriate engagement surface.

Mutual responsibilities

What we need from you

The schedule holds only when both sides can provide a small, specific set of inputs — named before implementation so the workflow does not stall at an authority or access boundary.

  • A named workflow. One specific process with visible friction, exceptions, or delay — not “an AI strategy.”
  • The operational owner. The person accountable for the work, available to validate the map and acceptance baseline.
  • The system owner. The person who can explain access, integration, and write boundaries for the systems involved.
  • The decision authority. The person who owns approval, release, exception, and stop decisions; security or compliance joins when the workflow requires it.
  • An operator. The person or team who will run the workflow and identify where the design fails under real use.

Engagement boundaries

What we don’t do

AI consulting is full of vague engagements that produce slide decks instead of operating change. We define the boundaries up front so each stage ends in a direct decision.

  • No staff augmentation. We do not place engineers inside your team without ownership of a bounded workflow outcome.
  • No generic AI strategy. If the engagement is not scoped to a named workflow, we do not take it.
  • No open-ended retainer. Every stage has a deliverable, acceptance boundary, and next decision.
  • No hidden authority expansion. Read, recommend, draft, write, approve, and release permissions are named before implementation.
  • No evidence laundering. Production observations, synthetic demonstrations, and roadmap claims remain visibly separate.

Pricing posture

Quoted against the workflow, not a generic price list

We do not publish tier pricing. Every engagement is quoted case by case against workflow complexity, system access, data boundaries, decision authority, acceptance criteria, and operating support.

What scopes the quote

Workflow count
primary scoping variable
Vertical pack complexity
framework count, jurisdictions
Evidence-boundary need
Tier 1 eligibility
Receipt volume
monthly request projection
Provider routing
Claude / GPT / Gemini mix

What you can expect

  • Workflow Opportunity Map — fixed deliverables, quoted schedule, output is yours
  • Accepted baseline — agreed measures, authority boundaries, and decision
  • Itemized implementation quote — modules, integrations, operation, and support
  • Outcome-based gates — named production milestones, not hourly billing
Map one workflow

Engagement questions

Questions buyers ask before they engage

How long does an engagement actually take?

The Workflow Opportunity Map is typically completed in two to four weeks. A bounded implementation may target six weeks when systems, access, and decision owners support that pace. Every schedule is confirmed against the mapped dependencies before the build begins.

What do we have to commit on our side?

One named workflow; its operational and system owners; the person who owns approval and stop decisions; access to the relevant systems and current artifacts; and an operator who can validate real exceptions.

Is this a pilot or a production deployment?

The Opportunity Map is a diagnostic. An implementation is defined by a named operating milestone and acceptance criteria, not by a demo. If the map does not support a build, the correct result is to reshape or stop.

What does it cost?

We do not publish tier pricing. Each engagement is quoted case by case against workflow complexity, system access, data boundaries, decision authority, acceptance criteria, and operating support. The Opportunity Map deliverables are yours regardless of whether you continue.

What happens if we stop working with you?

Handoff, continuity, and artifact portability are defined in the engagement scope. Any delivered offline-checkable sample or evidence export retains the documented verification method included with that artifact.

How do we know whether we are ready to start?

The AI Governance Readiness Assessment is a short, private self-assessment that scores your AI usage, governance maturity, evidence capability, and regulated-data exposure in about three minutes. It runs in your browser and is a signal, not a certification — a useful way to frame the Phase 1 conversation before you reach out.

Begin with the map

Begin with the Workflow Opportunity Map

One workflow, a measured baseline, an implementation brief, and a direct go, reshape, or stop recommendation. You keep the work either way.

Map one workflow