For enterprise
Working with us as an enterprise: inside your processes, not around them
Enterprise AI projects rarely fail on the technology. They stall in security review, get lost in procurement, or end as a pilot that nobody owns. We plan the engagement around those realities from the first call.
The people
Who is involved, on both sides
On your side
A sponsor who owns the outcome and the budget, a product or process owner for day-to-day decisions, and the IT, security and data contacts who review what we build. Often legal and procurement as well.
On ours
A named engagement lead for governance and reporting, a technical lead your architects can talk to directly, and the delivery team. The same people from the first assessment to handover.
Decisions
How decisions get made across many stakeholders
Decisions take longer in large organisations, and a good plan admits it. At the start we map who approves what (architecture, data access, security, spend) and build those review points into the timeline, instead of discovering them in week six.
Progress is reported against agreed milestones, with a written status every week and a steering review at the end of each phase. Changes go through your change-control process, with the impact on cost and timeline stated before anyone approves them.
Contract and payment
The shape of the deal
- Structure
- Usually an assessment first, then delivery in phases. Enterprise work starts at $10,000 and each phase is scoped and priced separately.
- Invoicing
- Enterprise engagements are invoiced monthly against an agreed schedule, to fit your finance calendar.
- Procurement
- We work through vendor onboarding and security questionnaires, and review your contract templates. Data-processing terms are agreed per project.
- Ownership
- Code, configuration, prompts and documentation transfer to you on final payment, deployed in your own cloud environment.
These are plain-English summaries. Your signed contract governs, and all terms are confirmed in writing for each project.
Honest risks
What usually goes wrong, and what we do about it
These are the failures we plan for, each paired with the thing we do to stop it happening.
Security review arrives late and stops the project.
Architecture and data-flow documents go to your security team in the first two weeks, so they review the design before anything is built.
The pilot never becomes production.
The first use case is chosen together with a named business owner, and the route to production is agreed before the pilot starts.
Access to systems takes weeks.
Accounts, data access and environments are requested in week one, and the plan front-loads work that can proceed while approvals move.
Knowledge leaves with the vendor.
Documentation, runbooks and handover sessions for your team are part of every phase, not an optional extra at the end.
The first 30 days
What happens after you say yes
The first month runs to a fixed shape: here is each stage, and what you get at the end of it.
Week 1
Kick-off and access
Kick-off with the sponsor and owners, a map of stakeholders and approvals, and access and environment requests submitted.
Weeks 1–2
Current-state review
Systems, data and process in scope reviewed; architecture and data-flow documents shared with IT and security.
Weeks 2–3
Proof of access
Read-only access proven, and a small technical proof of the highest-value use case on real data.
Day 30
Steering review
Findings, a recommended roadmap, a costed first delivery phase, and a go/no-go decision.
Where to go next
Transform
Where to start, what changes and what a roadmap covers.
AI integration
Connecting AI to the systems you already run, securely.
Enterprise package
From $10,000, scoped phase by phase.
Start with a scoped assessment
We reply within 24 hours on weekdays, and after a free call you get a written plan, timeline and fixed quote.