Thinking

Why most enterprise data challenges stem less from tooling limitations and more from misaligned definitions.

Why a proof-of-concept hides architectural fragility, and why the real work is making the ecosystem trustworthy at scale.

Why the right platform choice is rarely about technical features, and almost always about organisational alignment.

Why your GenAI proof-of-concept will fail in production without a boring, robust data foundation.

Why programmes don't fail dramatically at the end, but quietly during the initial assumptions phase.

Moving beyond copilots to build AI agents that fundamentally change how data engineering operates.

The difference between whiteboard designs and systems that survive budget cycles and team turnover.

How to embed data governance directly into the platform architecture instead of a Word document.

A framework for evaluating organisational maturity before evaluating software vendors.

Untangling fragmented estates requires clear migration pathways, not just a lift-and-shift mandate.