Custom AI workflows
Custom AI workflows, built around how your business actually runs
Most AI tools ask you to change how you work to suit the software. A custom workflow does the opposite: it's built around the way your business already runs, using your prices, your wording, your process — so there's nothing to migrate and nothing to learn.
- Built from your process, not a generic template
- Wired into the tools and data you already use
- You approve anything that touches money or customers
- Runs on our own hardware in New Zealand
What a custom AI workflow actually is
A workflow is a job your business does over and over: a quote gets written, an invoice goes out, an overdue account gets chased, a client gets an update, a document gets filed. Right now a person does each of those by hand, every time, and the cost is invisible because it's spread across the week.
A custom AI workflow is that same job, done by software that understands language. It reads the messy input — a voice note from the van, a photo of a job sheet, an email, a supplier PDF — turns it into something structured, does the work against your own rules and prices, and hands you the result to approve. You stay the decision-maker; you stop being the typist.
Why off-the-shelf AI usually disappoints
Generic AI tools are impressive in a demo and frustrating in a business, for one reason: they don't know anything about you. They don't have your rate card, your client history, your job records or your way of writing. So they produce confident, generic output that someone then has to check and rewrite — which is often slower than doing it properly the first time.
The fix isn't a better chatbot, it's grounding. A workflow we build is connected to your actual data, and the important rules live in code rather than in the model's head. That's the difference between an AI that guesses plausibly and one that's right — and it's why we build rather than resell.
How we build it
We start with a conversation about where your time actually goes — not a technology discussion. From that we map the two or three workflows with the best return, and build the first one so you can see it working with your own data before committing to more.
From there it's iterative: put it live, watch what it gets wrong, tighten it. Anything irreversible — sending an email, marking money as paid — asks you first and shows you exactly what it's about to do. That's a deliberate design choice, not a limitation: a system that confirms is one you can trust with real work.
Proof, not a pitch deck
We didn't just write about this — we built it
Quarvois our own platform, running a real New Zealand business every day: quoting, scheduling, invoicing, bank reconciliation and a full set of accounts, with an AI manager called Torq sitting across the whole thing. It runs on our own GPU hardware in New Zealand — so the data never leaves the country, and there's no per-token bill to a US API.
That's the difference between a consultancy that talks about AI and one that ships it. Everything on this page is something we've actually had to make work in production.
Questions we get asked
How is this different from just using ChatGPT?
ChatGPT is a brilliant general assistant that knows nothing about your business. A custom workflow is connected to your data, follows your rules, and runs on its own without anyone prompting it. The practical difference is that ChatGPT helps you do a task faster; a workflow means the task stops arriving on your desk.
Do I need to replace the software I already use?
Usually not. We prefer to build around what you've got — most businesses have systems that work fine, they're just surrounded by manual work. Replacement is a much bigger project and we'd only suggest it if what you're on is genuinely holding you back.
Where does my data go?
For most of what we build, nowhere — we run open-weight models on our own GPU hardware in New Zealand, so your business data isn't sent to an overseas API and isn't training anyone's model. Where a cloud model is genuinely the better tool we'll say so and explain the trade-off before anything is built.
What does it cost?
It depends on scope and we'll quote after a conversation, not before. What we can say is the shape: start with one workflow, prove it, expand. There's no large upfront project and no multi-year contract — if the first thing doesn't earn its keep, you haven't bet the business on it.
Talk to an engineer, not a salesperson
Ring either of us directly. You'll speak to the person who'd actually build it — no gatekeeping, no discovery call before anyone technical shows up.
Andrew
Founder & engineer
Builds the AI systems — including Quarvo, running a real NZ business every day.
021 149 5933Elliot
Engineer
Automation, integrations and getting it working inside your existing setup.
021 332 699
Rather write it down? Use the form below.
No bots · you talk to the developer who builds it
Start with a conversation
Tell us what your business does and where the time goes. We'll come back to you personally with what we'd build and whether it's worth it — no obligation, New Zealand based.