Eliaz Bobadilla

I started writing code at 14. Not tutorials — actual projects. That early habit of building to understand (instead of reading to understand) still defines how I work.

Over the past decade I've worked across early-stage startups, scale-up teams, and client projects spanning fintech, fitness, developer tooling, and infrastructure. Some of those shipped to thousands of users. Some got scrapped for the right reasons. All of them taught me something concrete.

My focus has shifted progressively toward backend architecture, AI systems integration, and scalable infrastructure — areas where engineering decisions made early compound hard in both directions.

I don't start with architecture. I start with the constraint. What's the real bottleneck — latency, cost, complexity, time to market? The answer changes the design more than any framework choice does.

I default toward simple and boring. That means reaching for proven tools before new ones, writing code that the next engineer can read without context, and treating every abstraction as debt until it proves its worth.

When AI is part of the stack, I treat it as an integration problem, not a research problem. The goal is reliable output, bounded behavior, and clear fallback logic — not a demo.