Services
A fixed-scope engagement when the problem is well defined. Advisory when it is not, by the project or on retainer. A senior practitioner leads every one.
Fixed-scope engagements
A defined deliverable and a set duration. The price is fixed after a short scoping call, before any work starts.
AI readiness assessment
An inventory of your data, candidate use cases and risk exposure, ending in a ranked roadmap with cost and payback for each item.
2 to 3 weeks
- You get
- An inventory of the data that candidate use cases depend on
- Candidate use cases, ranked
- A summary of risk exposure for each
- A roadmap with cost and payback for each item
- You provide
- A sponsor who can make decisions
- Access to the people who own the data and systems
- Existing architecture and policy documents, if you have them
Production agent build
One workflow, one agent, shipped to production with governance and evaluation built in from the first release.
6 to 8 weeks
- You get
- One agent running one workflow in production
- Policy checks and audit records from the first release
- An evaluation suite that runs on every change
- A runbook and a handover to your team
- You provide
- The workflow and the person who owns it
- Access to the systems the agent will touch
- Someone with authority to approve the policy
Model evaluation and red-teaming
A test suite for accuracy, drift and failure modes before launch, rerun on every model or prompt change.
3 to 4 weeks
- You get
- A test suite covering accuracy, drift and failure modes
- Red-team findings, ranked by severity
- The suite wired to rerun on every model or prompt change
- You provide
- Access to the model or endpoint under test
- Examples of good and bad outputs
- Your definition of an acceptable error
Advisory
Architecture and strategy for data, AI and cloud platforms: platform selection and RFP support, data architecture, cloud migration, and the operating model that keeps AI running after the project team leaves.
Typical questions we answer
- Should we build or buy our AI platform?
- Which pilots deserve production funding, and which should stop?
- How do we evaluate the vendor responses to our RFP?
- What governance will an auditor expect to see?
Advisory retainer
A C-level practitioner on retainer who works as part of your leadership team: setting AI and data strategy, reviewing what gets funded, and sitting in on the decisions that are expensive to reverse. For organizations that need that judgment every month, not for one project.
- You get
- A standing seat in your leadership or steering meetings
- AI and data investments reviewed before they are funded
- Architecture and vendor decisions checked before you commit
- Board and audit committee briefings when you need them
- A direct line between meetings
- Terms
- A set number of days each month and a minimum term, agreed on a call.