
RICE Framework for AI Prompting
I used to think AI was unreliable. I’d write a prompt, get something vaguely in the right direction, tweak it, get closer, tweak it again,
I work where business strategy, technology innovation, and architecture meet. I’m still a tinkerer at my core.
I’m an AI Architect and Product Architect at one of Europe’s largest energy companies. I build AI platforms, design cloud architectures on Azure, and help others understand what AI actually means for them. The practical reality, not the hype.
I built Kraken, an AI agent platform for one of the largest telecom provider in the Middle East, years before the AI and agentic systems hype. I created the RICE Framework for structured prompting. I’ve spent 15 years moving between mobile, web, cloud, and AI, from the Middle East to Europe, bridging the gap between technical teams and business leaders.
When I’m not architecting platforms or writing about AI, I’m experimenting with agentic systems, building open-source tools, and sharing what I learn.
If you’re figuring out where AI fits in your architecture, your team, or your strategy, let’s talk.

I used to think AI was unreliable. I’d write a prompt, get something vaguely in the right direction, tweak it, get closer, tweak it again,

While building an agent to answer some complex queries about my data, I noticed the first few drafts revealed something interesting. The screenshot below shows
Some cookies keep this site working. Others help us see how it is used. We only set the optional ones if you say yes.
Read our cookie policyNeeded for the site to work. These cannot be switched off.
Remember choices you make, such as your language.
Help us understand how the site is used. In the UK these need your consent.
Used to show you advertising that is relevant to you.