Trust centre

Built like an engineering notebook, not a hype machine

I write this content for people who need to deliver real outcomes. My goal is practical clarity and honest limits. If a topic is uncertain, I say so. If I have an opinion, I label it as an opinion.

This is a personal educational website. It is not affiliated with, endorsed by, or officially connected to my employer or any organisation mentioned.

Security

How to report a vulnerability and what I do with reports.

Privacy

What data is collected and how it is used.

Accessibility

How I aim to make the site usable for everyone.

Service status

High level availability updates for the platform.

How I keep content reliable

I use primary sources and recognised standards where possible.
I keep examples reproducible and I call out assumptions.
I update notes when frameworks, vendor guidance, or legislation changes.
Nancy AI runs local first with retrieval grounding, schema checked tool contracts, and policy gates before high impact actions.
I tune prompts and retrieval first, then run targeted fine tuning only when repeated gaps survive evaluator and contract tests.
I treat tool outputs as learning aids and I avoid pretending they are production truth.

Read updates Submit feedback

Tools are designed to be safe for learners

Many tools run in your browser and use fixed datasets or safe inputs. If a tool can create risk, it is labelled and it should be used on data you are allowed to use.

If you are ever thinking, "Is this safe to run on work data", the answer is usually no. Use a sanitised sample dataset and treat the tool like a lab, not a production system.

Verification and identity

If you want to verify who I am and what I do, start with the verification links on my About page.

View verification

A quick favour

If you spot something wrong or unclear, please tell me. A good correction is a gift. A great correction is free quality assurance.

Contact me Leave feedback

Preparing the page...