Capability
Agentic AI
Models given a bounded tool surface and a logged trail of everything they did.
Why us on this
Worth the cost where a process needs judgement across several systems and rules would never hold. Scoped narrowly, evaluated before trusted, and never the only path through a decision that matters.
Where we have run it
Public agency administering EU funding · internal audit · close to €1 billion a year in payments
On-Premise AI Internal Audit Assistant
An agency paying out close to €1 billion a year in EU funds cannot afford an audit finding missed because an auditor was on day five of retyping a regulation into a spreadsheet. Nor can it send that material to a cloud AI. So we designed an AI that never leaves the building and cannot cite a law that does not exist.
~80%Projectedless auditor time per control plan
Investment group · Europe · €150M sustainability-linked IFC facility with ESG covenants
ESG Compliance Intelligence Platform
A €150M sustainability-linked facility from the IFC rested on an ESG programme that lived in 70 scanned documents, 2,400 files and one person's inbox. The annual lender report took three months to compile by hand, and the IFC reads it with its own AI. We designed the platform that turns all of it into assigned, reminded, evidenced and auditable work.
€150MIFC facility the platform protects
Products in this capability
What we actually reach for, and what each one is good at.
Microsoft CopilotMicrosoft
Microsoft's AI assistant across Microsoft 365.
Self-hostable document chat and knowledge base.
OpenAI GPTOpenAI
OpenAI's large language models, used through the API.
Azure OpenAIPart of Microsoft Azure
OpenAI models inside the client's own Azure tenant.
Azure AI SearchPart of Microsoft Azure
Search and retrieval over the client's documents.
Anthropic's Claude models, used through the API and in our own engineering.
Ways to start on it
Fixed-scope engagements that run on this, priced before they begin.
Start here
AI Readiness Assessment
Three weeks to separate the processes where AI changes the economics from the ones where it is an expensive way to run a rule.
Duration3 weeks
Pilot
Agent Pilot: one process to production
Six weeks, one process, an agent that actually goes live behind guardrails, with the evaluation harness to prove it works.
Duration6 weeks
Lab
Emerging Tech Sprint
Two weeks testing one emerging technology against one of your real processes, ending in a recommendation you can act on or drop.
Duration2 weeks
Talk to us
Thinking about agentic AI?
Tell us where the work stops today. We will say plainly which tools fit it, what moving it takes, and when this is the wrong approach for what you are trying to do.