From a gap between systems to a production service.
The people who scope the work are the people who build it. We do not start with a platform pitch. We start with the work as it runs today — especially where it falls between systems that were never meant to talk.
What we take on
The work usually lives between systems: a spreadsheet next to an ERP, an inbox next to a GIS, a person reconciling two sources of truth. We map that gap, define what "correct" looks like, and put a production service in the middle — with monitoring, retries and an audit trail. Where judgement is genuinely required, the system asks a human, and learns from the answer.
We take two to five project assignments a year. If the gap is not frequent enough, expensive enough or well-defined enough to close, we say so at the first call rather than at the first invoice.
Step 01 — Find the gap
Two or three days with the people in the middle. We map what the large systems do not cover — volume, handling time, error rate — before proposing anything. If the gap is not frequent enough, expensive enough, or well-defined enough to close, we say so.
- Who does the work, how often, and which systems they currently touch.
- Where errors, delays and rework actually occur.
- Which parts require judgement versus which parts are mechanical.
- What “done” looks like, in a form a system can check.
Step 02 — Prove it on real data
A narrow slice in production within weeks, running alongside how the work is done today so the two can be compared directly. We do not demo on synthetic samples and call it a result. Accuracy, throughput and cost are reported against the current baseline.
Step 03 — Scale what holds
Widen the scope only where accuracy and cost hold up. Everything is infrastructure as code from the first commit. The system runs in your accounts, with EU and US data residency handled at the account boundary.
Step 04 — Hand over
The system ships as yours, under your name where that is the deal. Runbooks, dashboards and the code go with it, and we stay reachable for questions. We do not stay on as an operations team.
Own products and white label
Most of the year goes into products for our own companies. Most of those systems we then sell on to the organisations that need them most, under their name and in their accounts. Assignments are the smaller half of the business, which is why there are only a few of them.
- Own products — built for our own companies, to the standard we would want handed to us.
- White label — sold on to the organisations that need them. Your brand, your accounts, your operations.
- Selective assignments — two to five engagements a year, founder-led from first call to handover.
- Development, not ops — no on-call, no staffing, no managed operations contract.