How we work
The parts most suppliers leave vague.
None of this is unusual. It is written down because being specific about it in advance is the closest thing to a guarantee anyone in this field can honestly offer.
Engagements
One to two weeks
Technical discovery
A written assessment of whether the thing you want is feasible, what it would take, and where it will break. Delivered as a document you own and can take to another supplier.
You have a business case and need it pressure tested before committing budget.
Typically six to twelve weeks
Build
Scoped delivery against the discovery output. Fixed scope, working software, tests, documentation and a handover your own team can pick up.
The decision is made and you need it built by someone who has shipped it before.
Ongoing, monthly
Embedded
Working inside your team on your board. Code review, architecture, and levelling up your engineers so the dependency ends.
You have engineers but nobody who has taken an AI system to production.
Principles
Discovery produces a document, not a proposal
The output of the first phase is a written technical assessment you own outright. If you take it to another supplier and they build it, discovery still did its job. That is deliberate, and it is why the assessment is honest about the parts that will be hard.
Evaluation before features
Nothing ships without a way to tell whether it got better or worse. On an AI system that is not optional, because the failure mode is quiet: output that still looks plausible while being wrong more often than last week.
You own the code and the decisions
Repositories, infrastructure and accounts are in your name from day one. No wrapper platform, no per seat licence on your own process, and no situation where leaving costs you the system.
Handover is part of the build, not a phase after it
Documentation, runbooks and architecture notes are written while the work happens, because handover documents produced at the end are written by somebody who has already mentally left.
Bad news early
If an approach is not working you will hear it in the week it becomes apparent, not in the report at the end. The cheapest project is the one stopped in week two.
Tell us what you are trying to build.
Technical detail welcome. The more concrete the problem, the more useful the first reply.