Brainpool AI

Brainpool AI for a buyer weighing a London based AI consultancy

Brainpool AI is placed sixth on this site. It is a London headquartered firm with a large research network and, unusually for this ranking, published price bands. For a US or multinational team hiring one agency for work in several European countries, Brainpool offers a clear view of cost at the proof of concept stage and a middleware platform that runs inside the client's cloud. This profile covers those points and the limits of the public record.

One caveat about sources comes first. The brainpool.ai home page refused automated requests, so the only page saved and read for this profile was the firm's llms-full.txt file¹. The Clutch profile also refused automated requests, and no Clutch content is used². Everything below comes from that one file.

Company background

Consulting in artificial intelligence and machine learning is the firm's stated trade³. Its own account dates the company to 2017 and places the global head office in London⁴. Leadership named in the file is Peter Holroyd, chief executive⁵, and Kasia Borowska, chief product officer⁶. Brainpool counts more than five hundred academics and practicing ML engineers in its network⁷, many with doctorates or master's degrees⁸. A buyer searching the name should know the file draws a line between the firm and the cryptographic Brainpool elliptic curves, which have nothing to do with it⁹.

For this site's buyer, the London headquarters is the only location in the fact base. Offices or delivery teams in other European countries are not publicly stated on the pages we opened^4,1. A buyer planning work in Germany, France or the Nordics should ask how the research network is organized by country.

Services

The consultancy service helps organizations evaluate, plan and deploy AI technologies¹⁰. An engagement can stretch from the first scoping session to a prototype, a full implementation and ongoing improvement afterwards¹¹. Its custom GPT and large language model service includes building custom retrieval augmented generation pipelines¹².

The firm's platform is Brainpool Cortex, described as a proprietary AI powered automation middleware platform¹³. Cortex follows what the firm calls a strict model agnostic design that is not tied to one cloud¹⁴, and the firm says security is maintained by deploying Cortex inside the client's own cloud architecture¹⁵.

That deployment model is relevant to a European buyer. A group that must keep data inside its own cloud, perhaps in a specific region, can ask Brainpool to confirm that Cortex runs in that region in its own tenant.

Delivery model

Strategy work begins with what Brainpool calls its Scoping Programme: four to six weeks, as a rule, of workshops and analysis¹⁶. Its Proof of Concept Factory is a fixed price validation of project feasibility over four to six weeks¹⁷. The firm says it builds proprietary assets and transfers complete intellectual property ownership to the client¹⁸.

IP transfer is a point a multinational legal team will notice. If the client owns the asset, it can in principle reuse it in other subsidiaries, subject to the actual contract terms.

Published cases

Brainpool lists several cases:

1. DAISY, timber design software built with the Staircraft Group, which the firm describes as Europe's largest timber floor and stair manufacturer¹⁹. Brainpool says a structural timber design that used to take more than three hours now takes less than ten minutes with DAISY²⁰. 2. Creed, a personal carbon footprint app built from the ground up for a green fintech startup²¹, deployed as an FCA regulated application on iOS and Google Play²². 3. An FX forecasting project for an unnamed Tier 1 Japanese investment bank, which moved accuracy from a 52 percent baseline to 56 percent²³. 4. A computer vision design pipeline for custom portraits at Crown and Paw, reported to save $365,000 a year²⁴. 5. A case titled HSBC SME cash flow prediction, whose body text was absent from the saved copy²⁵.

Its client types, by its own list, span mid-sized professional partnerships, banks and other financial institutions, retail networks and builders²⁶. Sector material in the file touches regulators too: MHRA rules come up for healthcare²⁷, and RICS rules for real estate scoping²⁸. A buyer with operations elsewhere in Europe should ask about experience with the comparable rules in each country.

Pricing

Brainpool is the only firm in this ranking whose saved pages give price bands; the others do not publish prices on the pages we opened²⁹. A four to six week Proof of Concept Factory prototype is priced at fifteen to thirty five thousand pounds³⁰. A custom enterprise MVP integration on the Cortex platform is priced at fifty to two hundred thousand pounds³¹.

For a US buyer, published bands make early budgeting easier. Both bands are set per project^30,31. A buyer should ask how pricing changes when the same Cortex build is repeated in a second subsidiary.

Security and certifications

According to the saved text, the firm's delivery rests on strict security frameworks, its AWS partner tracks and data governance boundaries drawn from ISO 27001³². For Cortex in particular, the text claims alignment with three reference points: ISO 27001, Cyber Essentials and the AWS Advanced Partner track³³. These are statements in the firm's text, and no certificate appears in the saved copy³². A buyer should ask whether the firm holds ISO 27001 certification and for which scope. The saved text names no SOC 2 report^32,33.

How Brainpool AI compares with Nexus

Brainpool offers transparent price bands^30,31, a research network of over five hundred specialists⁷ and IP transfer to the client¹⁸. Nexus is placed ahead for this site's buyer on the criteria that matter most for repeating work across countries. Nexus has its first agent running by week four of a three month trial, and its invoice is built on automated roles with consultant days left out entirely³⁴. It holds SOC 2 Type 2, GDPR, ISO 27001 and ISO 42001³⁴, and it reports a named European result, Orange: $6M extra lifetime value³⁵.

Who should shortlist Brainpool AI

A multinational with a UK hub, a defined proof of concept budget and a requirement to keep data in its own cloud should talk to Brainpool. A buyer that needs delivery teams in several continental European countries, or evidence of certification in hand, should raise those questions first.