What we build
We scope narrow on purpose. If a process is unstable, or an off-the-shelf tool already covers it, we say so on the audit call rather than after the invoice. Every engagement is fixed price against a written scope, delivered into your own repository.
A WhatsApp number that answers, qualifies and books without a person reading every message first.
Wrong for you if a human has to calculate a quote before any useful reply can be sent.
Agents scoped to one job with defined inputs, a defined output, and a human checkpoint where the cost of being wrong is high.
Wrong for you if the process has no written rules and no history of past decisions to learn from.
Self-hosted on your infrastructure, versioned in your repository.
Wrong for you if it is a simple two-step integration — a stock SaaS connector is cheaper.
New bots where the rules are stable, and rescue work on bots that keep breaking when a UI changes.
Wrong for you if the process changes monthly — maintenance will cost more than it saves.
Invoices, claims, KYC packets and contracts turned into structured data — with a confidence threshold and a review queue for everything below it.
Wrong for you if the data already arrives as a feed — an API, EDI or database export.
Moving the brittle parts of an existing estate to AI models while leaving the parts that work alone.
Wrong for you if you have fewer than five bots and they are working fine.
Pricing bands
Published because you should not have to sit through a call to find out whether we are in your range.
A single process, scoped narrow, quoted at a fixed price with a delivery date in writing before work starts.
Several connected processes across one department. Typical delivery is four weeks per process, sequenced rather than run in parallel.
Optional by design. Thirty days of defect fixing is already included in every build, and nothing we ship depends on a retainer to keep running.
Typical delivery is four weeks. Model and messaging usage is billed to you directly by the provider, with no margin taken by us. Every engagement is quoted against one process, one price and one date, written down before work starts.
Straight answers
You do, directly to the provider. You use your own AI API key and receive your own bill — we take no margin on tokens and we do not resell usage.
You do. Code goes into your repository, on your cloud, with your credentials, from day one — not at handover.
Yours. The WhatsApp number and the Meta Business Manager stay in your name throughout, so nothing is stranded with us if you stop working with us.
Documentation and a runbook are standard deliverables, and thirty days of defect fixing is included. Ongoing support is optional by design, never a dependency.
Yes, and we turn work down when maintenance would cost more than the saving. You get that answer on the free audit call rather than after the invoice.
No. We self-host n8n on your infrastructure, so there is no per-task metering from us and your bill does not climb with volume.
Start with one process
Forty-five minutes, no charge, no deck. You will get a straight answer on whether it is worth automating — and if it isn't, you will get that answer too.