Question 1

How do engagements usually start?

Accepted Answer

With a paid discovery: a two-week process trace where we watch the work happen, record every step and exception, and produce a map of what to automate, what to keep human and what to delete. You own the output whether or not you continue with us.

Question 2

How long does a first release take?

Accepted Answer

Most engagements put a working vertical slice in production within six to eight weeks of kickoff. Full programmes run three to six months, delivered as complete workflows rather than a single launch.

Question 3

Do you work fixed-scope or time-and-materials?

Accepted Answer

Both. Discovery is always fixed. Delivery is usually time-and-materials with a capped budget per phase, because scope on AI work shifts as evaluation results arrive.

Question 4

Who owns the code and the models?

Accepted Answer

You do — code, prompts, datasets, evaluation sets and infrastructure definitions, in your repositories and your cloud accounts, from the first commit.

Question 5

What happens after launch?

Accepted Answer

We offer an operate agreement covering on-call, drift monitoring, evaluation maintenance and a monthly improvement cycle. Clients who prefer to run it themselves get a structured handover including documentation and paired on-call.

Question 6

How do you handle data protection and compliance?

Accepted Answer

Data handling is agreed before any data moves. We can design systems with privacy, data residency and security requirements in mind, and work with clients to meet the requirements appropriate to their use case, including signed DPAs where required.

Question 7

Which models do you use?

Accepted Answer

Whichever the evaluation supports. We build model-agnostic and route per task — a small model for classification, a larger one for reasoning — with cost and accuracy measured per route.

Question 8

Do you replace our existing systems?

Accepted Answer

Rarely. Most of our work integrates with the EHR, PMS, CRM or ERP you already run. We replace spreadsheets and manual handoffs far more often than we replace platforms.

Question 9

Can you work with our in-house engineering team?

Accepted Answer

Yes, and it is our preferred model. Embedded delivery with your engineers in the pairing rotation produces a better handover and a lower long-run cost.

Question 10

What size of client do you work with?

Accepted Answer

From multi-site clinics and regional field service networks to mid-market platform teams. The common factor is an operation with volume, not a particular headcount.

The questions we are asked before every engagement.

Let's work on something that has to work.

The questions we are asked before every engagement.

01How do engagements usually start?

02How long does a first release take?

03Do you work fixed-scope or time-and-materials?

04Who owns the code and the models?

05What happens after launch?

06How do you handle data protection and compliance?

07Which models do you use?

08Do you replace our existing systems?

09Can you work with our in-house engineering team?

10What size of client do you work with?

Let's work on something that has to work.