Work

What AI actually looks like doing real work.

Not slideshows or someday-maybe ideas — tools quietly running inside a real business every day. I'm showing you these so you can picture what's genuinely possible for a team like yours, once the hype is set aside.

A few examples

Each one started as a problem someone needed solved.

Finance · multi-entity group

Executive financial dashboards

Live accounting data across a multi-company group pulled into one place — profit & loss, cash position, aged debtors, budget versus actual — with an AI layer that explains the numbers in plain language. Read daily by leadership.

Built configuration-first; figures reconciled against source before anyone quotes them.

Operations · voice capture

Voice-to-publish staff pipeline

Staff publish updates just by speaking. AI cleans and formats the text while preserving the speaker's own voice; every item then passes a human approver before it goes live. In production with around thirty-five staff.

Human-in-the-loop by design — nothing publishes itself.

Knowledge · grants

Grants assistant

Reads a grant's structure from a link or uploaded form, drafts a first-pass application grounded in the organisation's own knowledge base, then scores that draft against the funder's criteria — a person in the loop at every step.

Used end-to-end to put a real funding application out the door.

Operations · inventory

Inventory system of record

An append-only stock-movement ledger with running average cost and deterministic production-run costing, feeding straight into quarterly financial reconciliation. Replaced a spreadsheet held together with hope.

One source of truth the finance system can actually trust.

How I work

Built carefully, so you can trust it.

01

Real, not a demo

Everything is checked against what's actually running — not a flashy prototype. And if something isn't proven yet, you'll hear that plainly.

02

Honest about limits

AI is genuinely useful and it also gets things wrong. I'll always be straight with you about where it helps and where to be careful.

03

You stay in control

A person stays in the loop on anything that matters. AI does the heavy lifting; the decisions stay with you and your team.

You don't need to be technical
to get real value from this.

None of the tools above needed a developer. They needed someone willing to describe a problem clearly and work through it patiently — which is a skill you already use every day. That's the part I can help you and your team find your way into.

Wondering what this could look like for your team?

No pressure and no commitment — if you're curious, I'm happy to talk it through and tell you honestly what's worth doing.

Let's have a chat How I help teams →