How an engagement works.
We agree what needs to change, build the systems with your team and hand them over in accounts you control. You can run them independently or choose our ongoing service.
We build it with you, and can keep running it afterwards.
Our engineers use AI development tools to build and test your systems. You review working versions on your own data, with agreed checks before production.
- Implementation
- We connect your data, map the business rules and rebuild the workflows that are hard to rely on. You review working versions throughout.
- Optional ongoing service
- A named engineer and an AI assistant work with you in a shared Slack channel, monitor your workflows and make agreed improvements.
- Expansion
- Add workflows and data sources to the same foundation. Your team describes a change in plain English, and an engineer tests it before release.
The first milestone
Working software comes early in the programme.
Programmes run over 30, 60 or 90 days, and useful parts go into use as they are ready. The ten-day plan below covers the first milestone only.
Book a call, opens the 20-minute booking page in a new tabAn example first milestone
- Before day1
Agree the programme. Set milestones, fees and acceptance checks. Confirm the access, data and baseline needed to start.
- By day3
Review a working version. Try the first workflow on your own data. We test the normal path and the exceptions together.
- By day10
Put it into use. The first workflow handles real work, with approvals, monitoring and a named owner. The wider implementation continues.
Binding dates and refunds are set in the agreement for each milestone.
Dates start once the agreement, access and data are ready.
How we keep it running.
We replace fragile chains of automations with tested, monitored workflows, and an optional named engineer keeps them running.
- Monitored execution
- Jobs and integrations are monitored, and failures alert the person who fixes them.
- Recovery when something fails
- Workflows retry and resume after an interruption without repeating finished work.
- Changes checked before release
- Changes are tested against your business rules, and important actions keep an approval history.
- Capacity
- We test for expected volumes and plan capacity as usage grows.
You own the system and can keep building on it.
PATCH, our engineering foundation, connects your data and business rules with the workflows and AI agents that use them. You can also commission the data foundation on its own.
- Your code and database
- The repository, Postgres database and services run in accounts you control, with documentation to operate and develop them.
- Open to inspection
- Code, data, rules and execution history stay open to your team, with a record of what ran and why.
- Changes in plain English
- Describe a change and an engineer tests it against your rules before release. New workflows build on what is already there.
- Independent of us
- Keep developing it with us, your own team or another provider. Licences and running costs are listed in the agreement.
Who you work with.
You work directly with the people responsible for your implementation and any ongoing service.
Enterprise
For larger and regulated organisations.
We work with your operational, technical and security teams to define the systems you need. Data access, deployment, approvals and delivery stages form part of the implementation plan.
Our work includes connected business data and query systems for Beluga Health. Data handling and regulatory requirements are agreed for each engagement, and we do not hold a security certification.
Enterprise programmes have their own milestones, testing and approval process. The ten-day plan is not an enterprise delivery commitment.
