Building Sovereign AI from organisational knowledge.

The knowledge behind trustworthy AI.

I turn manuals, SOPs and technical documents into structured, verifiable knowledge, with extracted text, tables and figures checked against their original sources.

Answers that always show their source.

I build assistants that answer from an organisation's own approved documents, using knowledge models and retrieval designed to give grounded, source-cited answers.

What a system should and should not answer.

I define where an assistant's knowledge ends, how it handles uncertainty, and where a person needs to review an answer before it is relied on.

Knowledge that stays firmly in-house.

I design systems that run on locally hosted models, so an organisation's documents and intellectual property stay on its own hardware, with governance and human oversight designed in.