Build products
Ecommerce, management apps, complete frontends. The goal was to make them work and deploy them. I learned to finish things.
Criterion: does it work?
Full-Stack Developer
I am Luciano Carrizo, a Full Stack Developer in Buenos Aires. I build end-to-end products —backend, data, frontend— and for the past two years I have also built the tools I use to build them.
It is not a colorful side note. Those are the first two things the court taught me, and I later recognized them under different names inside a repository.
I started at 14 at Club Atlético San Miguel and have played for IJCP since 2019. This year we practice three times a week, plus match day: for more than ten years, my week has been organized around it.
Tuesday practice does not depend on how Tuesday feels. Keeping three practices a week for ten years teaches something unglamorous: what is yours to do gets done. I treat delivery commitments the same way.
One player finishes the point, but reception and setting came first. On court, the credit is never one person’s alone. That transfers directly to work: what matters is that the play succeeds, not who finishes it.
Ten years of practice are mostly repetition with nobody watching. The work nobody sees is what sustains a match. Tests, documentation, and conventions are my version of that.
You reach Metropolitana’s second division one step at a time, not in a single leap. Each division shows you what you did not yet know you were missing. Spending time on the step below, learning, is part of the method—not an accident.
Ecommerce, management apps, complete frontends. The goal was to make them work and deploy them. I learned to finish things.
Criterion: does it work?
I started paying the cost of my own decisions. I separated layers, isolated the domain, wrote tests that proved something, and documented the reason behind each choice.
Criterion: can I change it in six months?
The next bottleneck stopped being code and became context. I built persistent memory, agent orchestration, and reusable working disciplines across projects.
Criterion: can I delegate and verify it?
Each phase absorbs the previous one rather than replacing it. I still deliver products; what changed is the level at which I make decisions.
Every architecture choice records its discarded alternative and its reason. If I cannot remember a year later, the repository can.
Clean Architecture where it pays off: business logic does not know the framework, database, or transport. Infrastructure changes without touching the rules.
I run the affected flow and observe the real behavior.‘The test ran and passed’ is not verification—it is evidence.
Conventions, solved bugs, and decisions live beyond the session. Context does not disappear when I close the terminal.
One task per agent, explicit input, verifiable output. AI accelerates execution; design and review are not delegated.
READMEs, specs, and technical case studies a third party can follow without asking me anything. It is the same discipline I use to instruct an agent.
Six years of electromechanics before information systems. The pivot was not about abandoning a degree: it was realizing that what interested me about a system —how it behaves, where it fails, what makes it maintainable— was better solved in software.
A formal foundation in data structures, databases, and software engineering—where project intuition becomes named judgment.
Two years of calculus, physics, and systems modeling. They remain with me: thinking about constraints, tolerances, and failure modes before features.
Six years of workshop practice: diagnose before touching, measure before claiming. Real verification as a reflex, not an optional step.
HTML, CSS, and JavaScript from scratch: semantic layout, responsive design, and end-to-end website logic.
Link unavailableThe language in depth: types, functions, the DOM, asynchronous work, and patterns that later appear in every framework.
Link unavailableComponents, state and props, hooks, and routing—the foundation for building a real SPA.
Link unavailableFrom raw data to decisions: cleaning, exploratory analysis, and asking the right questions before charting anything.
Link unavailableDashboards and data stories: connecting sources, calculations, and visualizations designed for someone else to understand.
Link unavailableBeyond the spreadsheet: advanced formulas, pivot tables, and simple models for organizing real data.
Link unavailableComplementary education was a quick entry into the stack. Judgment came later, from the workshop and university.
If you are evaluating profiles, start with the code. If you would rather talk, the contact panel is one shortcut away.
or directly → LuchoC.dev@gmail.com