Some teams need to decide, build, or operate an AI system. Others need independent judgment or better adoption of tools already in place. Each engagement stands on its own, with scope and price agreed before work begins.
Start a conversation →The AI Decision Sprint. A fixed fee of $15,000–$40,000 over 2–4 weeks that identifies which opportunity is worth funding and returns a build or no-build recommendation with a costed path to production.
Explore →Design, integration, evaluation, and delivery into production on your stack. Fixed-fee phases, typically $75,000–$150,000 each, with written acceptance criteria and an explicit stop or go decision at the end.
Explore →A defined monthly service ($5,000–$15,000) that keeps a production system accurate and adopted: evaluation monitoring, optimization, governance updates, and a bounded allowance for improvement work.
Explore →Objective, dependencies, stakeholders, and the acceptance criteria that define done, agreed in writing before the phase begins.
A published range, then a fixed price per phase before work begins. Sprint fees are credited in full toward the first Build phase within 90 days. New scope becomes a new agreed phase, not a meter.
Every phase ends in a deliberate decision. Continuing is a choice, not a default, and stopping is a legitimate outcome.
Not every organization needs Celadon to build something. Two services exist for the ones that do not: one for the leaders making the decisions, one for the people expected to use the tools. Neither is a phase in the progression above, and neither requires a Celadon build.
A retainer of $5,000–$12,000 / month for leaders who want senior, vendor-neutral judgment on architecture, supplier decisions, evaluation methodology, and investment priorities while their own teams do the work.
Explore →A fixed-fee program of $20,000–$60,000 that turns AI your company already licenses into daily practice: role-specific playbooks, a policy people read, and usage measured per team against a baseline.
Explore →| If you need to… | Best fit | What changes |
|---|---|---|
| Choose where AI deserves investment | Decide | A ranked opportunity, business case, and build or no-build recommendation. |
| Deliver a selected system | Build | A production system with evaluation, integration, documentation, and acceptance criteria. |
| Maintain the technical health of a live system | Operate | Measured quality, governed changes, system usage signals, and a bounded improvement cadence. |
| Give an internal team independent judgment | Advisory | Vendor-neutral decisions on architecture, suppliers, evaluation, and priorities. |
| Change human workflow and adoption | Adoption | Role-specific playbooks, workable policy, enablement, and value measured by team. |
A live system can need both Operate and Adoption: one maintains technical health; the other changes how people use the tool and realize value. Neither silently includes the other.
Send the problem or opportunity you have in mind. We will reply by email with the right starting point, or tell you honestly if AI is not the answer yet.
Tell us what you are evaluating, building, operating, or trying to get adopted. Email is the first step; if a call would be useful after we review the context, we will suggest one.
