Seven ways we take work off your team.

You do — all of it, from day one. We don't build on a proprietary platform you'd have to keep renting from us.

• Code lives in a repository owned by your company. We're contributors, and we can be removed.
• Cloud, database, and AI provider accounts are created in your name, so the providers bill you directly — nothing runs through us.
• Your data stays in your accounts. We never hold it hostage, resell it, or use it to train anything.
• Domains and DNS stay with you. We'll ask for temporary access, not ownership.

If you part ways with us tomorrow, everything keeps running and another developer can pick it up. That's deliberate.

Security is a design constraint from the start, not a review at the end. Practically, that means:

• Least privilege. Every integration gets the narrowest access that works. No shared admin logins, no keys in spreadsheets.
• Encryption in transit and at rest on every system we build, using your cloud provider's managed keys.
• Data minimisation. If an AI assistant doesn't need a patient's full record to book an appointment, it never sees it.
• No training on your data. We use enterprise API endpoints where customer data is excluded from model training by contract, and we'll show you that clause.
• Audit trails on anything that touches customer records, so you can answer "who changed this?"
• Offboarding discipline. When someone leaves your team or ours, access is revoked the same day.

If you're in a regulated sector — health, finance, legal — tell us early. We'll work to your obligations and put them in the scope document, and we'll sign your NDA rather than asking you to sign ours.

A period of tuning is included in every build — that's when reality finds the gaps.

After that you choose:

• A support agreement. A named contact (a person, not a queue), a regular review of the numbers, and an agreed response window for issues. Month to month, cancellable anytime.
• A clean handover. Documentation, a recorded walkthrough, and your in-house person or IT provider takes it. We'll answer their questions through the transition.

We don't lock support to a 12-month contract. If we're only kept because leaving is expensive, that's a bad relationship.

First: things we build degrade rather than fail. An AI assistant that loses its connection stops answering and routes to your team with a message, instead of guessing or going silent. Automations retry, then queue, then alert a human — they don't drop work quietly.

• Monitoring on every system we run, alerting us before your customers notice.
• A documented manual fallback for anything customer-facing, so your staff know exactly what to do while it's down.
• An escalation path on support agreements — answered by the person who built it.
• A written post-mortem for anything that reached your customers, including what we changed so it can't repeat.

And if something breaks because we built it wrong, we fix it — that's on us, not you. That's not a favour, it's the job.

Fair question, asked honestly. Here's the honest answer.

We're a good fit if you're an established business running between roughly 5 and 200 staff — clinics, professional firms, trades, multi-branch operators — where a system needs to work reliably for real customers and real money.

We're the wrong choice if you need 24/7 follow-the-sun coverage, a dedicated on-site team, or enterprise procurement with a 40-page security questionnaire and a nine-month cycle. We'd rather tell you that now than fail slowly.

What being small actually buys you: senior people on your project instead of whoever's on the bench, decisions made in a day, and no margin funding a sales floor. We keep active engagements to a handful so nobody's work becomes filler.

Where we need more hands — a specialist integration, video, clinical review — we bring in a named specialist and tell you who they are and why they're on it.

It starts with a free diagnose session — a call or a working meeting, no charge and no obligation. We find where your hours, leads, or margin are actually leaking, and what's worth doing about it.

• From there you get a clear written proposal for the specific work: what we'd build, in what order, and what it should return.
• Nothing is committed until you've seen that in writing and said yes. No rate cards, no pressure.
• Third-party services (cloud, AI usage, licences) sit in your own accounts, billed to you directly — they never run through us.

And if the numbers don't justify the build, we'll say so. A project that can't pay for itself is usually the wrong project.

Usually no. Starting over is expensive, slow, and often unnecessary — and it's the reflex answer from anyone who bills for rebuilds.

We work with what you have wherever it's sound. If your CRM is fine but nothing talks to it, we connect it. If your site converts but looks dated, we fix the pages that matter. We recommend replacing something only when the cost of working around it is higher than the cost of moving off it, and we'll show you that arithmetic.