Chile · Solutions Architect · Engineering Lead
Solutions Architect · Engineering Lead
I design systems that scale. I lead teams that understand why.
About
+15 years making technical decisions that affect real systems. From migrating a monolith to microservices across 5 countries, to co-founding a platform that reached 50,000 students. Today I lead the regional modernization of MiLocal at Cencosud. When I'm not doing that, I take on consulting and development projects where the architecture problem is genuinely interesting.
How I Think
Real principles, not Twitter maxims.
· The domain does not know AWS exists
Ports and adapters isn't a pattern for tutorials — it's the difference between replacing a storage provider in a day or in three weeks. If your business logic imports an infrastructure SDK, you don't have architecture: you have coupling with good presentation.
· All debt gets documented. None gets hidden.
I've taken on technical debt consciously and defended it to stakeholders. I've also inherited debt nobody knew existed — that's what paralyzes teams and delays releases. The difference isn't having debt; it's knowing exactly where it is and what it costs.
· Not everything needs to go through your API
Excessive control has a cost: latency, bandwidth, and unnecessary operational complexity. Designing flows where the client interacts directly with the right service is not losing control — it is understanding the responsibility boundaries of each component.
· Consistency first, elegance second
A separate events collection alongside the aggregate looks like correct DDD. Until the aggregate write fails and the event is already saved — no alarm, no rollback, silent inconsistency. The atomic write is less elegant on paper and more reliable in production. I choose production.
· AI multiplies speed. It does not replace judgment.
I evaluate and combine models by decision type: Claude for architectural design, Gemini to stress-test approaches, Copilot to accelerate implementation. What I don't delegate is judgment. Using it without foundations only accelerates poor decisions. The judgment is still yours.
Experience
Cencosud / Manpower – Experis
The problem wasn't technical — it was that nobody knew the real size of the debt. My job: make the invisible visible, prioritize with business judgment, and execute without breaking what already works across 9 countries.
Manpower / Experis for Cencosud
Two critical systems, two countries, one architecture that had to adapt without breaking. Rompe Filas handles POS payments with different gateways per country. Altillos optimizes storage logistics with complex hierarchies. The result was good enough to earn me the transversal role.
Amaris Consulting — Esmax/Aramco, Cencosud
Critical systems in Azure with DDD and Event-Driven. The real challenge was expanding e-invoicing to 3 countries with different regulations without touching the core — that's where hexagonal architecture paid for itself.
Codelovers
9 years. 200 schools. 50,000 students. What starts as an MVP in CodeIgniter ends as a migration to Symfony, then AWS, then React. ClassTrack survived because the architecture was designed to change, not to last.
Stack
Certifications
AWS
AWS Solutions Architect – Associate
AWS
AWS Developer – Associate
AWS
AWS Cloud Practitioner
GitHub
GitHub Administration
GitHub
GitHub Foundations
GitHub
GitHub Advanced Security
GitHub
GitHub Actions
GitHub
GitHub Copilot
MongoDB
MongoDB: Relational to Document Model
Writing
Articles on architecture, technical decisions, and systems that matter.
Read on Medium →Consulting
Facing a complex migration, accumulated technical debt, or need a senior architectural second opinion? I work with teams that already have developers but need senior judgment.
Projects where the architecture problem is genuinely interesting. Not corporate websites — systems with real logic, complex integrations, or scale that matters.