Workflow automation
Turn repetitive intake, routing, and review into a clear, auditable flow.
AI systems for teams where the work has consequences
I design and run AI workflow automation for legal, accounting, and other knowledge-intensive teams — from the first bottleneck to a dependable production system.
Book a working session →Human approval by design · EU-conscious delivery · Built for production
What I build
The goal is not to add another tool. It is to remove a recurring bottleneck without giving away control.
Turn repetitive intake, routing, and review into a clear, auditable flow.
Use models where they help — drafting, classifying, and finding the next action — with people at the decision points.
Ship the infrastructure, observability, and handoffs that make an AI workflow dependable.
Specialist services
Each one is a standing offer with its own page, not a bullet on a capability deck.
AI literacy training and EU AI Act readiness for teams that have to answer for the tools they use.
Explore enablement →02Track how AI assistants cite you, and run the content pipeline that keeps you in the answer.
Explore visibility →03MCP servers, evaluation harnesses, and the document and ERP plumbing underneath.
Explore integration →Products
Useful automation should be easy to understand, easy to evaluate, and safe to own.
Engagements
Every engagement starts with a 20-minute fit call. API and platform usage is billed at cost.
How we work
Find the expensive, repeatable part of the work and define what good looks like.
Prove the workflow with your rules, your data boundaries, and your team in the loop.
Measure what happens in production, then expand only where the evidence says to.
Built and running
I build done-for-you AI automation for teams drowning in manual work — and I publish the numbers, including the parts that failed.
Inbox and intake automation: a 10,000-email morning turned into a manageable stream.
A daily publishing pipeline: draft, human review, publish, distribute.
A public paper-trading experiment with every number traceable and seven failed ideas documented.
LLM extraction through to auto-published market commentary.
Built for careful work
The right automation makes the next action clearer, while keeping responsibility visible.
Automation handles the first pass. Your team keeps the decisions that matter.
Data access, sending permissions, and escalation points are part of the design.
The system earns trust through observable, reviewable behavior — not a pitch deck.
We will map what is worth automating, what should stay human, and what a sensible first step looks like.
Book a working session →