AI is the newest layer, not the beginning of the story
Why AI does not replace the fundamentals of architecture, data, users and systems.
WRITING · NOTES · REFLECTIONS
What I learn
What I learn
while building.
Thoughts on technology, systems, artificial intelligence, airports, products and how ideas become real things.
Not a content treadmill.
A place to keep ideas worth developing.
A place to keep ideas worth developing.
WHY WRITE?
Building produces answers.
Building produces answers.
Writing helps explain why.
I write to preserve the ideas, decisions, mistakes and lessons that appear while doing the work.
ARCHIVE
Notes and essays.
Functionality before cosmetics
Why a product that works matters more than an impressive demo that falls apart in the real world.
From eight-hour prompts to model orchestration
What almost four years of working with AI taught me about tools, models and efficiency.
What airports taught me about systems
When thousands of people depend on a system in real time, technology stops being theoretical.
When an AI-generated MVP becomes harder to fix
Why quickly generating a prototype can sometimes move the cost into debugging, architecture and maintenance.
Saving 10% of tokens wasn’t enough
A model-orchestration experiment that showed why someone else’s results do not automatically transfer to your own constraints.
Not every problem needs AI
Sometimes a better interface, database or simpler workflow is the better answer.
Why I build prototypes to understand
Building as a way of learning: question, prototype, break, learn and rebuild.
The difference between a demo and a system
A demo proves an idea can work. A system has to keep working when reality gets complicated.
Problems often start with the data
Why fragmented data often creates more friction than a lack of tools or automation.
RECURRING THEMES
The things I keep coming back to.
AI
Systems
Building
Airports
Customer Experience
Architecture
Accessibility
Data
Product Thinking
Experiments
A SIMPLE IDEA
The tools change.