Ranges from past work, and an estimator that narrows them to your scope. Your project could cost a lot less than the top of a range; every quote follows a written scope.
Introduction transcript. Client: How much does an AI project cost? Qubi: Let’s measure the job before we make a promise. Qubi: These ranges are a starting point. Your quote follows a written scope. Client: No mystery extras hiding in the lining? Qubi: We spell out the assumptions. And you own the code. Qubi: Tell us what you need. We’ll work it out together.
Prove an idea, choose a direction, build for production, or keep improving after launch. Every one-off range is drawn on the same € scale, so you can see how they compare.
Indicative ranges from past client work. All prices are in EUR, exclusive of VAT. Your quote follows a scoping call and a written scope. These are estimates: many projects can be scoped to cost less.
Estimate yoursFor proving a technical approach works — investor demos, internal buy-in, or de-risking an architecture before committing to it.
€8–25k
For deciding what to build. Assessment of data, systems, constraints and feasibility, with a costed shortlist.
€8–20k
Building the system properly: integrations, evaluation, monitoring, error handling and handover.
€30–120k
Fractional capacity after launch — quality monitoring, iteration, and building the next thing alongside your team.
€4–9k/ month
What each engagement is for, what you leave with and how it is priced. Pick one to bring its column forward.
| Criterion | Prototype Sprint | Discovery | Production Build | Ongoing |
|---|---|---|---|---|
| Best for | Proving an approach before you commit to it | Deciding what to build, and in which order | Launching a system people rely on | Improving a system after launch |
| You leave with | A working system on your data and a written production assessment | A costed shortlist and a reference architecture | An integrated, evaluated and monitored system, handed over | Agreed days each month, monitoring and model migrations |
| Typical range | €8–25k | €8–20k | €30–120k | €4–9k / month |
| Typical duration | 2–4 weeks | 2–4 weeks | 6–16 weeks | Month to month |
| How it is priced | Fixed quote | Fixed quote | Fixed for a bounded scope; phases for larger work | Monthly capacity, month to month |
| Natural next step | A production build, or a clear no | A build with us, or with another vendor | Ongoing care, or fully in-house | In-house, once your team is ready |
An agent can mean one automated step or work across your entire stack. These are three common shapes, estimated from past work and quoted on their own scope.
One job, one or two tools. Triage inbound email, draft replies from a knowledge base, enrich CRM records, summarise and route tickets.
€8–25k
An agent that carries a role: support, sales research, data analysis or internal ops. Retrieval over your documents, several integrations, escalation paths.
€25–75k
Several agents coordinating across ERP, CRM, email and your data warehouse, with approvals, audit trail and a compliance review to pass.
€75k+
Tell us what you are building and how. The figure moves inside the ranges above, shows its working, and can land below them when your team takes on part of the work.
€41–59k
7–9 weeks · EUR, excluding VAT
Arithmetic on your answers over the ranges on this page: not a quote, and not a measured client outcome. Your project could cost less. The written quote is what counts.
| Item | Your answer | Effect |
|---|---|---|
| Starting point | Lowest figure in €30–120k | €30k |
| Systems it connects to | 2 systems | +€10k |
| State of your data | Some clean-up needed | +€9k |
| Who uses it | Your team | — |
| Accuracy bar | A person checks the output | — |
| Compliance | Standard | — |
| Who does the work | We build it | — |
| Central figure | Likely €41–59k | €49k |
Every answer adds a share of the published range above its lowest figure. Integrations carry the largest share, each system more than the last, then the state of your data. Answer everything at its most demanding and you reach the top of the range; at its simplest, its floor.
Your engineers taking a share of the work scales the whole figure down, which is how a project can come in under the published range. The likely range runs from 15% under the central figure to 20% over it, because scoping always finds something.
Build spread evenly across its months for illustration; real invoices follow agreed milestones. Model inference, hosting and tools are paid by you, on top.
Real shapes of past work, with client names left off, on one € scale. Slide your budget across them to see what it covers.
€40k covers the whole range of 2 project shapes, and the lower end of 4 more.
One workflow on real data, deployed to a URL, written production assessment
Retrieval over your documents with citations, permissions respected, evaluation set
Extraction and classification of invoices, contracts or forms into your systems, with human review queue
Six-week build of the core feature, one or two integrations, ready to put in front of users
Ticketing and CRM integration, escalation paths, guardrails, monitoring and cost tracking
Agents acting across three or more systems (ERP, CRM, email), approvals, audit trail
Laravel build, design system, structured data, llms.txt and lead capture
Several systems, several phases, compliance review, dedicated capacity
Indicative ranges from past client work, in EUR excluding VAT. Your scope can land outside these ranges, below as well as above. The written quote is what counts.
Six things we look at before estimating the work, and which way each one moves it. Integrations and data readiness usually have the biggest effect.
| What we look at | Brings it down | Pushes it up |
|---|---|---|
| 01 Integrations Biggest effect | One system with a clean, documented API can be about a week of the work. | Three legacy systems with undocumented behaviour can be most of the budget. Each additional system tends to add more than its share. |
| 02 State of the data Biggest effect | Clean, accessible, labelled data keeps a project at the low end of its range. | Scattered, inconsistent or permission-locked data pushes it up before any AI work begins. |
| 03 Who sees the output | An internal tool used by ten people is cheaper than the same tool facing customers. | External users bring guardrails, escalation paths, monitoring and a higher bar for evaluation. |
| 04 Accuracy bar | A person reviews the output before anything is acted on. | Getting from good to reliably good is where effort concentrates. The acceptance criteria agreed in discovery decide how much evaluation the build needs. |
| 05 Compliance and security | Standard data on your usual hosting. | Regulated data, on-premise or EU-only hosting, security questionnaires and audit trails all add real work. They are costed rather than absorbed. |
| 06 How much you do | A client engineer who owns data access and deployment can take a meaningful share in-house. The project gets cheaper and the handover easier. | We carry every part of the work ourselves, from data access to deployment. |
The estimator above weighs the same six things, in the same order.
The figures above describe what past clients paid. The proposal describes your project.
A free, usually 30-minute scoping call to understand the systems, data and constraints. We say early if the budget and scope do not meet.
Agree what will be delivered, the assumptions and acceptance criteria. A structured Discovery engagement is paid work with its own deliverable.
A fixed figure for bounded work, or agreed phases and capacity for a larger programme. You can review the scope before committing.
For well-bounded work, the figure stays fixed unless the scope changes. If it changes, we re-quote at the boundary. The estimating risk sits with us.
Multi-phase or open-ended work runs on phases or agreed capacity per month, with a fresh estimate at each phase. Several systems, teams or a compliance review call for a bespoke proposal.
We do not offer a blanket accuracy guarantee with a refund attached. A useful commitment needs defined outputs, an agreed sample and a way to judge the result. Those details belong in the statement of work.
During discovery, set acceptance criteria against a held-out evaluation set.
If a milestone misses the agreed criteria, we fix it at no charge.
If we cannot meet those criteria, we agree a fair reduction for that milestone, under the statement of work.
We do not charge per seat, per token or per agent. Our fees cover the agreed work; inference, hosting and third-party tools remain your running costs. These ranges are estimates, not a rate card: integration effort and data preparation become clear during scoping.
Our ranges sit between offshore agencies and marketplace freelancers, and the larger consultancies. This is our positioning, rather than a published market price comparison.
For similar work, they tend to quote below us.
No account managers and no bench to keep busy, so we can price the work rather than the overhead.
For similar work, they tend to quote above us.
We do not publish other suppliers’ figures without a source you can check, so this scale has no numbers on it.
All prices here are in EUR and exclude VAT. Spanish IVA is charged where it applies; EU businesses with a valid VAT number are reverse-charged.
No. They are estimates drawn from past client work, published so you get a realistic idea of cost before we talk. Every project is different and is quoted on its own scope after a scoping call. Once quoted, well-bounded work is a fixed figure against a written scope; larger or open-ended work is priced in phases.
Yes. The ranges describe typical past work, not a floor. A narrower first version, one integration instead of three, clean data, or your own engineers taking a real share of the work can all bring a project in under them. Tell us the budget and we will say what fits.
It is arithmetic on your answers over the ranges published on this page, and it shows every step. It is a sensible place to start a conversation, not a quote: integration effort and data preparation only become clear during scoping. The written quote after a scoping call is the figure that counts.
Because the alternative wastes both our time. If your budget is €10k and the work is a €60k build, a call establishing that is a call neither of us needed. Hidden pricing often ends up being priced against perceived budget rather than against the work, and we would rather not start a relationship that way.
Then none of these bands apply and we will say so. A large programme — several systems, several phases, compliance review, dedicated capacity — gets a proposal written for it, usually priced phase by phase with a fresh estimate at each one. Trying to squeeze it into a published band would give you a number that is confidently wrong.
Say so early and we will tell you honestly whether the scope can be cut to fit. Sometimes it can — a narrower prototype, one integration instead of three. Sometimes it cannot, and then we would rather point you toward an off-the-shelf product or a smaller supplier than take an underfunded project that disappoints everyone.
Roughly €8–25k for a single-workflow agent with one or two integrations, €25–75k for a production agent that carries a whole role with retrieval, several integrations and escalation, and €75k upwards for a multi-agent system priced in phases. Those are estimates from past work; the number of systems the agent has to act in and the accuracy bar you need are what move it. Running costs — inference, hosting, tools — are separate and paid by you.
For ongoing and open-ended work, yes: the Ongoing model is a set number of days per month, and phased programmes can run on agreed capacity. For bounded projects we prefer a quoted figure, because it puts the estimating risk on us rather than on you.
No. The first call is free and usually 30 minutes. Discovery — the structured assessment with a written deliverable — is a paid engagement, because it is real work producing something valuable whether or not you go on to build with us.
Yes, and they are yours rather than ours: model inference, hosting and storage. We model these during the build so there are no surprises, and cost optimisation is part of the work rather than an upsell. If you take the system fully in-house, you owe us nothing further.
We do not do equity-for-services or revenue share. It aligns incentives in theory and in practice creates a conflict about scope on every subsequent decision. Quoted fees keep the relationship clean, and we would rather be a supplier you can fire than a shareholder you cannot.
We will tell you within a call whether they meet — including when they do not.
30-minute call · no deck
What we built, what stage it is at, and a link to the live thing wherever there is one.
yourOS is an operating system with an AI at its heart. You talk to it like a person, out loud or typed. It makes your calls, plans your days and works on your projects with a crew of agents, on your own private server. Our own product, built from the first commit; the waitlist is open.
Website, backend and app assets for Sonora, a free AI sound-therapy app on iOS and Android published by Cognitive AI.
Brand, website, prototype, apps on five platforms, the AI backend and the algorithm-intelligence pipeline for Podswap, a social growth engine that runs without a human in the loop.
In closed beta: Roxy, a private AI assistant with permanent memory running on infrastructure the user owns. Brand, site, product, backend and conversational design.
Live at coherence.com: brand identity generated with AI agents we directed, the motion system and website for the group behind Sonora, AboutAI, WakeUp and MudShadows. Agent work remains in build.
Live at domainbooth.com for DomainBooth, a premium domain brokerage: a new animated mark, every page redesigned and rebuilt with designed mobile layouts, and a custom CMS for its sales record, featured names, press and enquiries.
Apps, platforms and campaigns for these teams, long before the AI work. Every one is written up on the work page — what we did, and where our part ended.
