ENGINEER. ARCHITECT. AUTHOR.
Zul Mehdi.
Enterprise thinking.
AI-native architecture.
I’m an engineer-turned-architect building AI and cloud systems for the enterprise. I connect the big picture to the details that make it work in production.
Author of the RICE Framework for AI Prompting. Writing about the systems I build and the lessons behind them.
Malmö, Sweden
01 / THE FRAMEWORK
CREATED BY ZUL MEHDI
The RICE Framework
for AI Prompting.
Better AI output starts with clearer thinking. I created RICE to turn an ambiguous request into a structured prompt: Role, Intent, Context, and Expectations.
Explore riceframe.work- 01R
Role
Who should the AI be?
- 02I
Intent
What should it achieve?
- 03C
Context
What does it need to know?
- 04E
Expectations
What does good look like?
02 / FIELD NOTES
Ideas from the work.
Architecture, AI agents, and the occasional experiment. I write to make sense of what I’m building.
Agent-Driven Self-Healing and Auto-Ticketing by a Multi-Agent Incident Response Team
An agent-driven closed-loop self-healing and auto-ticketing workflow, where LLMs interpret incidents, Jev guides bounded decisions, and the team keeps the record current.
Do AI Agents Really Need Tickets?
Do AI agents need tickets to begin work? Separate the execution brief from the human coordination, accountability, and traceability that ticketing provides.
My AI Agents Have Their Own Development PC
I gave my AI agents their own development environment on a shared Raspberry Pi. What started with a dedicated Linux account and Rootless Docker became a lesson in the difference between reduced privilege and real isolation.
03 / EXPERIENCE
From engineering to architecture.
A career grounded in building systems. Now focused on the architecture that lets AI work at enterprise scale.
// RECENT PROFESSIONAL EXPERIENCE
AI, product, and cloud architecture — with hands-on platform experience underneath.
Full career on LinkedIn- CURRENT
AI Architect - Nordic
🇸🇪E.ON
Cloud Architect (Sr. Consultant)
🇸🇪Klarna
Senior Engineer
🇦🇪Etisalat
Developer Support Analyst
🇦🇪Microsoft
Developer Support Analyst
🇦🇪Nokia
04 / PRACTICE
The big picture. The working system.
- 01
AI architecture
Designing multi-agent systems that hold up in production: clear layers, explicit contracts between agents, memory and evaluation built in from the start rather than bolted on.
- 02
Cloud architecture
Platform and landing-zone design for large enterprise estates — cost, security and operability treated as first-class requirements, not an afterthought.
- 03
Technical leadership
I've led engineering teams as tech lead, translating between business intent and system design, and helping teams make decisions they can still defend a year later.
- 04
Writing & frameworks
I write about what I learn on the job, and turn recurring patterns into things other people can use — like the RICE framework for prompting.
05 / THE PERSON
Beyond the diagram.
I grew up around technology in Dubai and now build it in Sweden.
Today I work as an AI and cloud architect across cloud platform, AI and engineering programmes. Before that I spent years as a tech lead, which is where I learned that the hardest part of architecture is rarely the diagram — it's keeping a team aligned on why the diagram looks the way it does.
In my free time I write at zulmehdi.blog, mostly about multi-agent systems, enterprise architecture and the cloud. Writing is how I think things through — and if it saves someone else a few weeks, even better.
If you're working on something in this space, I'm always happy to talk.