Design and delivery of retrieval-augmented generation systems over your own documents, data and knowledge bases, with evaluation built in from day one.
Answers grounded in your own documents, with a citation for every claim
Measured retrieval accuracy instead of demo-day impressions
A system your team can extend without calling me back
Turn "it seems better" into a number. I build evaluation harnesses, labelled test sets and CI gates so you can change an LLM system without breaking it.
Every change measured before it reaches users
Regression caught in CI, not in production
A defensible quality record for auditors and regulators
LLM agents that use tools, call your APIs and complete multi-step work, with the guardrails, observability and human checkpoints to deploy them safely.
Multi-step work completed end to end, with a human checkpoint where it matters
Every action traced, replayable and reversible
Clear boundaries on what the agent may do without approval
The infrastructure under your AI features: gateways, prompt versioning, tracing, cost controls, caching and pipelines that let several teams ship safely.
One governed path to production for every team building with models
Cost visible and attributable per team, feature and customer
Model changes and provider outages absorbed without incidents
A short, evidence-led engagement that tells you which AI use cases are worth building, which are not, what each will cost, and in what order to do them.
A ranked shortlist with honest feasibility and cost attached
A clear no on the use cases that will not work, with reasons
A sequenced roadmap your board can actually approve
from €22,000
How I work
How do you price engagements?
Fixed price for a defined scope wherever possible, quoted after a short scoping conversation, with the assumptions written down. Where scope cannot be fixed up front, such as an exploratory discovery or an audit of a system I have not seen, the work is time-and-materials against a capped estimate.
Can you work alongside our in-house team?
Yes, and it is the shape that works best. I embed with your engineers, build together, and hand over the evaluation harness, architecture decision records and runbook. The measure of a successful engagement is that your team makes the next change without me.
Do you offer ongoing support after a build?
A light retainer covering evaluation review, model migration advice and architecture questions, typically a day or two a month. No managed services, because a consultant who depends on running your system has the wrong incentives when designing it.
What is the smallest engagement you take?
A two-day architecture or evaluation review, which suits teams who have built something and want an honest second opinion before scaling it. Below that, a 30-minute call is free and often enough.
Get a straight answer on your AI roadmap
A 30-minute call with the engineer who would do the work, not a salesperson. You will get an honest read on what is worth building, what is not, and roughly what it costs.