Hernán Valencia

I have spent seven years building software that other people depend on during their working day, and most of what I believe about engineering comes from having to maintain it afterwards.

I work as a consultant on US telecommunications and insurance systems, and independently with startups and small companies that need a senior engineer across both the application and the platform. The work has ranged from designing new products to reducing a legacy container image from three hundred known vulnerabilities to none, without changing what the application did for its users.

My technology choices are deliberately unexciting. I pick tools by how well a team can operate them a year from now, not by how new they are, and I would rather deliver a smaller system that holds under load than a broader one nobody can debug. When a legacy system is genuinely carrying a business, I look for the incremental path — API extraction, strangler migrations, environment parity — before proposing a rewrite.

I like being close to the business context. Knowing why an invoice takes two days to process, or why an approval sits waiting for a signature, usually changes the design more than any framework decision does. That is also why I am comfortable in client-facing technical conversations: the decision has to survive both the code review and the budget review.

Outside of client work I keep a Fedora-based development environment, write the engineering notes published on this site, and spend a fair amount of time reading incident write-ups from teams operating at larger scale than mine.