José Silvaaka Zé Pedro

Collector of metaphors, mapper of systems, essayist at blog.zepedro.com. Born in Portugal, living in Ireland, citizen of the world.

About

I grew up in Portugal and now call Dublin home. My path from writing my first lines of code to guiding architecture across large-scale distributed systems has been shaped by a simple realization: the hardest part of building software is almost never the code itself—it’s how systems, teams, and human habits fit together.

I'm that person who still gets a genuine kick out of an elegant refactor, the quiet satisfaction of deleting 2,000 lines of dead code, or sketching an architecture diagram that makes a sprawling problem feel suddenly obvious. I believe great architecture isn't handed down from an ivory tower; it’s about clearing away friction so engineers can build with momentum and pride.

These days, I spend much of my time thinking about evolutionary platforms, distributed resilience, and what the rise of AI tools means for the craft of engineering. When machines can generate infinite syntax in seconds, human discernment, taste, and architectural judgment matter more than ever.

Writing is how I untangle my own thinking. Because software architecture is invisible, I rely on metaphors—wild gardens, police sketch artists, cartographers who stopped traveling—to give complex ideas a tangible shape. When I’m not sketching systems or writing on blog.zepedro.com, you’ll find me tinkering, drinking far too much espresso, and enjoying life on a rainy Dublin afternoon.

Essays

Notes from the climb: essays on systems, teams, and the human side of building.

Recent Dispatches
The wrong map: why the best question is a guess you expect to be corrected

Why showing an expert your draft understanding gets better answers than asking them to explain

↗
Hard or wrong: how to tell the difference when you’re already committed

Most teams are good at starting hard bets. The discipline worth building is knowing how to evaluate one while you’re still inside it.

↗
The political capital ledger: why some engineers go broke and never know it

Why being right doesn’t make a withdrawal free, and how the engineers who object to everything end up heard by no one

↗
The forger’s apprentice: what a police sketch artist knows that AI doesn’t

What a sketch artist and a witness can do that neither can do alone, and why the gap gets worse, not better, once nothing reviewing the code has eyes

↗
The second grid problem: what the microservices era teaches us about building in the age of AI

What the microservices era taught us about proliferation, ownership, and the hidden cost of building everything from scratch

↗
The gym membership principle: why defining fitness functions isn’t the same as using them

The specific ways fitness functions fail go beyond neglect, and some of them look like diligence

↗
The cartographer who stopped traveling: how familiarity hides the friction in platforms you built

Building platforms is an act of mapmaking. The question is whether the map still describes the territory.

↗
The test drive question: why curious questions transform technical reviews

Why groups of smart engineers collectively ask shallow questions, and what it takes to build a room where the real inquiry survives

↗
The garden that grows back wild: why resilience decays the moment you stop practicing it

How drift into failure, organizational forgetting, and the quiet physics of entropy turn well-built resilience into an illusion

↗
The borrowed sense: why you can’t borrow AI’s engineering instinct

What AI coding tools generate, what they can’t transfer, and how engineers at every level can tell the difference

↗
Elsewhere

Connect, collaborate, and read across the web.