Five steps to ensure successful AI implementation.
Every AI implementation runs through five stages. It starts with a free 30-minute intake, then a one-week audit (€670, credited against the project) that maps the highest-value place to put Claude. From there comes design, then deploy, then a handoff where your team is trained and owns the system. You get a fixed price before any building starts, and at the end the system is yours to run.
A short call. We decide together whether Brinvik is the right fit before either of us invests a week.
You leave with a yes or a no, and the price of the next step. If it is a no, you owe nothing and you know where to go instead.
I map how your team works and where Claude earns its keep. You leave with the plan.
You own the assessment either way: a map of how your team works, where Claude pays off first, and a fixed price you can act on or decline.
Prompts, skills, integrations, and the guardrails around them. Written down.
The whole system on paper before anything is built. Prompts, skills, integrations and guardrails, written so your own team can read them.
I ship it. HubSpot, Calendly, Notion, Slack, whatever your stack needs.
A working deployment in your own environment and your own tools, tested against the numbers we agreed in Design.
Your team owns the system. I document it, train two operators, and leave the keys. Retainer optional, not assumed.
You get two trained operators, the documentation and the keys. Nothing runs through Brinvik unless you choose the retainer.
A short call to decide together whether Brinvik is the right fit, before either of us invests a week. You leave knowing yes or no, and what the next step costs. It usually starts on brinvik.com via the agent.
This is the AI readiness audit. I map how your team works today, find where Claude pays off first, and write up scope, timeline, and a fixed price for the full engagement. You get a documented assessment you own either way, and the audit fee is credited against any project that follows.
I turn the audit into a buildable plan: project structure, skills, integrations, data flow, and the success metrics we will hold the work to. You review and approve it before any building starts, so there are no surprises in the build.
I build the system, connect it to your stack (HubSpot, Notion, Slack, Calendly), and validate it against the success metrics we agreed in Design. You get a working deployment in your own environment, tested against the numbers, not a demo.
I train your team, document the system, and turn over admin access, so the people who use it can run it without me. After handoff you either operate it independently or move into a Retainer. Either way, the system is yours.
Ongoing optimization, new deployments, monthly review. For teams that have decided Claude is part of how they run. Detailed on the Retainer service page.
See the Retainer →Where do yours go? Pick your role and I show you what I usually take off the desk.
You pick a slot in Kim's calendar. No form, no callback.
No logos. The tools I use don't make the work better. The work makes the tools earn their keep.
Focus produces depth. Multi-model is the consultant's hedge against having a thesis. Brinvik is a thesis: Claude's reasoning quality and the way it behaves with serious B2B work make it the right model for businesses that want one.
That focus shows up in the work. I know what Claude is good at and what it isn't. I know which integrations matter. The studio doesn't spend half the engagement deciding which model to use. The focus is the moat.
Tell me what you're trying to ship. It starts with a one-week audit and a fixed-price quote you can act on or walk away from. Either way, you leave with a plan.
You pick a slot in Kim's calendar. No form, no callback.
Certified by Anthropic · Fixed price on every stage · Invoiced in DKK
