I work on a question that has followed me from research into entrepreneurship:

How can AI make people better at making decisions, without asking them to surrender their judgement?

​

My career began in machine-learning research, including work in healthcare, where the limitation of a highly accurate model was often obvious: it was not enough for a system to produce an answer. Doctors, researchers and organisations also needed to understand it, challenge it and decide when to trust it.

​

That lesson became more concrete when I moved from research into building enterprise software.

 

As Co-Founder and CTO of Lapis AI Studio, I helped build an AI platform for regulated financial institutions. The work reinforced something I now believe deeply: the hardest part of Enterprise AI is rarely the model itself.

 

It is the system around it.

​

It is how knowledge is structured; how permissions are enforced; how calculations can be reproduced; how decisions can be traced; how security, reliability and human workflows are treated as first-class engineering concerns. It is the difference between an impressive demonstration and a system an organisation can depend on.

 

This site is where I develop those ideas in public.

 

I write about trustworthy Enterprise AI: the engineering, governance and product decisions that make advanced systems useful in the real world. I am especially interested in the point where probabilistic AI meets deterministic software; where human expertise meets machine scale; and where organisations need systems that are not only capable, but understandable and accountable.

 

I do not believe AI should replace human judgement.

 

I believe it should extend human capability.

​