How I work
Four convictions. Each one is backed by something that actually happened.
Fix the ceiling first
Every technology estate has a constraint that caps everything built above it. A brittle data layer. An integration nobody owns. A delivery process nobody governs. Add capability on top, whether that is AI, a new platform or a bigger team, and the ceiling rises on paper and stays exactly where it was in practice.
I find the constraint, fix it and then build. In a regulated, 26-country business the governed data platform came first. Three production AI systems followed.
No AI theatre
AI theatre is a pilot that demos well, a steering group that meets monthly and no line in the P&L that moved. Production AI has an owner, governed data behind it and a number it has to beat.
Every AI investment I recommend is assessed by commercial return before anyone spends. The RAG tool I delivered cut engineering information requests from sales by 45%. That is what a result looks like.
People decide whether it lands
Technology and platforms matter. I sincerely believe that what really moves the needle is the positivity of the team delivering them. It creates a can-do culture, where challenges are hurdles and not obstacles, and it keeps a team moving when things do not go to plan.
As Head of Engineering I built and mentored a delivery-focused development function around positive, high-accountability habits. For technology hires, positivity is non-negotiable. Talent is talent, regardless of experience.
Buy before you build
Bespoke software is a decision that needs a business case, not a default. In pharmaceutical logistics I chose off-the-shelf tooling wherever bespoke was not justified, and the estate shipped across 26 countries.
Start with the problem, not a job title.
Tell me what is happening, what is coming and what nobody in the business can currently answer.