Engaging with us
How do engagements usually start?
With a two-week AI Discovery, or a shorter scoping call if you already know what you want built. Discovery produces a ranked shortlist of use cases with effort, value and risk against each, plus an honest list of the ones to drop.
Do we have to buy the whole three-stage programme?
No. Discovery and Blueprint stand alone, and plenty of clients take the Blueprint and build in-house. You decide at the end of each stage whether to continue.
How small a project will you take?
Small enough to be useful, large enough to be worth both parties' time. A four-week POC is a normal first engagement. A two-day script is not — we will point you somewhere better.
Do you work outside your own region?
Yes. We deliver across five countries, working in the client's time zone for standups and reviews. Data residency is handled per contract, not per office location.
Pricing
How is the work priced?
Three models: a fixed fee for discovery and well-defined builds, a monthly team rate for embedded engineering, and a monthly retainer for managed infrastructure. Whichever applies is quoted before work starts.
What does a discovery cost?
A fixed fee based on the number of workflows and stakeholders involved, agreed in the proposal. It is not discovered halfway through.
Are infrastructure costs separate?
Yes, and they are quoted as a monthly figure sized to your actual usage before the build starts. On Stellar Cascade there are no per-request or egress charges.
Is there a per-seat AI licence?
No. Adding users does not change what you pay us. Infrastructure scales with usage, which is a different and much flatter curve.
Ownership and lock-in
Who owns the code and the architecture?
You do — code, architecture, models and documentation. There is no runtime licence to us and nothing that stops you taking it elsewhere. That is a deliberate constraint on how we design.
Will our data be used to train a model?
No. Your content is not used to train any shared or public model. Where a use case needs fine-tuning, the resulting weights belong to you and stay in your environment.
What if we want to leave mid-project?
You keep everything produced to that point, including the decision log explaining why things were built the way they were. There is no exit fee and no termination penalty beyond notice.
Do you use proprietary frameworks?
We build on standard, widely supported tooling. If we ever propose something unusual, the Blueprint says why and what the alternative would cost.
Security and data
Where does our data live?
In your own cloud account, your own hardware, or a dedicated single-tenant Stellar Cascade region in the country you choose. Nothing reaches a public model API unless you ask for that and sign it off.
Can you sign our NDA and DPA?
Yes, and we review rather than rubber-stamp. Where a clause is not something we can honestly commit to, we say so before signing rather than after.
How do you handle access to our systems?
Named individuals, least privilege, time-boxed access and your own identity provider where possible. Access is revoked at the end of an engagement as a checklist item, not on request.
What about prompt injection and model misuse?
Retrieved content is treated as data, never as instructions. Systems carry guardrails, output filtering and logging, and red-teaming is part of the build for anything customer-facing.
Delivery
How fast can we start?
Discovery usually starts within two to three weeks of a signed proposal. If timing is genuinely urgent we will say whether we can do it properly at that speed, rather than agreeing and under-delivering.
Will you work inside our codebase?
Yes, that is the default. Our engineers work in your repository, branching model and review process, at your sprint cadence.
Who will actually be on the project?
Named engineers, introduced before the contract is signed. We do not sell senior people and staff juniors, and the team does not change silently mid-engagement.
How do you report progress?
A weekly written update against the agreed milestones, plus whatever ceremony your team already runs. If something has slipped you hear it that week, not at the review.
After launch
What happens once it is live?
Either handover with runbooks and training, or we continue to run the platform under an SLA. Both are common, and the choice is made before the build finishes rather than after.
What does the SLA cover?
Availability targets, response and resolution times by severity, and a named engineer on call. Consequences for missing them are in the contract, because an SLA without consequences is a brochure.
Who retrains models as things change?
Your team, using the workflow we hand over. A system only your vendor can update is a system that goes stale, so retraining is part of handover rather than an upsell.
Do you offer ongoing improvement work?
Yes, usually as a small monthly allocation alongside the managed retainer. It is capped and visible rather than open-ended.
Question not answered here?
Send it over. An engineer replies within one working day, and if it is a common one it ends up on this page.