staff engineer · 25 years in production · AI-native development
Engineered
with AI.
I’ve shipped software that couldn’t fail: for 20M customers, for aircraft, for satellites. Then 13 production apps in 70 days, solo, with AI agents. Now I bring that discipline to how teams build with AI, so the agents multiply your engineers instead of flooding them with slop.
how i build
Build the right thing. Engineer it to hold. Lead the agents that write it.
01
Build
Decide what to build, and what to leave out. I settle the shape in a spec, before a line of code. That is the first call, and the hardest.
02
Engineer
Make it hold under real load. Kernel-grade rigor and compliance. Production systems, not prototypes.
03
Lead
Direct the workforce. I led teams across four markets; now I direct AI agents. Set the contract, own the result.
selected work
Systems that shipped and stayed shipped.
iFood: Scaling AI Adoption Across 3,500 Engineers
First member of the DATA & AI department in Marketplace — the largest business unit at Latin America's leading food delivery platform. Capturing, connecting, and packaging AI patterns that emerge organically across 30+ squads.
view project →
Crypto Fintech: 13 Apps in 70 Days
A full crypto payment platform (gateway, OTC exchange, on-chain settlement) engineered solo in 70 days with AI agents. 13 apps, ~1,650 tests, production-grade with compliance.
view project →
TUI Group: Checkout for 20M+ Customers
Tech Lead for the global checkout system at the world's largest tourism company (€23B revenue). 12-person team, 4 markets, +40% reliability, promotion recommended.
view project →
writing
Notes on building software in the age of AI.
the method
Building the AI-native way?
I write about AI-native development every week: the discipline behind shipping with AI agents. Follow the work, or reach out.