What's included

Subagents that perform complex tasks (e.g. coding) require so-called “frontier” models, which are billed separately at $15 per million tokens. A million tokens is a lot of text; see current flagship Anthropic (Claude) model rates in the Claude pricing documentation.

Tokens for normal usage of the system are included up to 100M tokens — an allowance that normal usage would not typically exceed. Beyond that, usage is billed weekly at $50 USD per 100M tokens.

Exact technical scope is fixed in writing after kickoff; the three bespoke items are chosen to fit the one-off fee and timeline. Need more headroom from day one? Start on the Independent track instead.

How your foundation package runs

From engagement through setup, then appraisal on live usage, and finally how you can grow into ongoing build or broader access.

Your timeline

Engagement

Scope & commitment
01

Kickoff & deployment scope

We meet to review the capability list, align on the three bespoke outcomes, and write down exactly what will be in the first deployment — so there are no drifting expectations later.

02

Custom feature definition

We name the three main problems the platform must solve or materially reduce in this first state — whether that is automated handoffs, specialist subagents for repeatable work, or a specific throughput bottleneck.

03

Scope agreement & payment

Written scope, milestones, and the single one-off fee are confirmed. You complete payment online and we queue the build.

Setup & onboarding

Deploy & walkthrough
04

Platform setup deployed

We stand up your environment with the full foundation — assistant, knowledge base, integrations, and dashboard — and put it in a state you can log into and trust before bespoke work lands.

05

Features deployed

We implement the three agreed capabilities on top of that foundation and wire them into real workflows — so what you see in the product matches what we signed up to build.

06

Onboarding with stakeholders

Key people are walked through registration, first tasks, and how the bespoke features show up day-to-day — so adoption starts confidently, not with guesswork.

Use and appraise the system

Appraisal
07

Monitoring & refinement

Automated telemetry plus human review show what is earning its place and what is not. We tune instructions, agents, and automations against real usage for as long as you need to appraise the system — not against lab assumptions.

08

Review cycle

Structured checkpoints on outcomes, friction points, and what should change next. You leave each cycle with a clear read on what is working and what is not before you commit to the next chapter.

Grow the system to your needs

Grow the footprint
09

Active development package

Move onto an active development track at £1,900 per month (Independent) for ongoing build, support, and iteration when you want hands-on development as a steady rhythm rather than a single engagement.

10

Collaborator package

Upgrade to Collaborator when you need colleagues or third parties in the same workspace — shared projects, fine-grained permissions, and weekly check-ins — and when you want to sell access to your platform to your own clients. See Collaborator on our packages page.

11

Group package

For multiple businesses with enterprise, siloed data requirements operating from the same platform, the Group tier provides the isolation and governance you need across organisations on one stack. See Group on our packages page.

Want a named platform before you commit monthly?

Book a call and we'll sanity-check fit — whether the foundation package is the right entry, or whether you're already a candidate for full ongoing development.

Questions? Contact us directly

Discovery call Our packages