macOS tar destroys files on Linux: I validated it in my real Railway pipeline and documented the 3 cases nobody mentions There's a Hacker News thread that resurfaced this week with 107 points about a 2024 article: tar on macOS creates archives that Linux can't extract cleanly. The community reacted the way it always does — "use GNU tar", "install gtar with Homebrew", "this has been known for yea
Tar en macOS destroza archivos en Linux: lo validé en mi pipeline real de Railway y documenté los 3 casos que nadie menciona Hay un hilo en Hacker News que resurfaceó esta semana con 107 puntos sobre un artículo de 2024: tar en macOS crea archivos que Linux no puede extraer limpiamente. La comunidad reaccionó como siempre: "usá GNU tar", "instalá gtar con homebrew", "esto es conocido desde hace
When you automate backups, you eventually discover the backup was not the hard part. The hard part was everything around it. This week I got a nice little reminder from my self-hosted agent setup: the backup job can be logically correct, authenticated, scheduled, and still fail because of two very boring constraints: Docker-owned files are not always readable by the user running cron. GitHub Relea
Something shifted in the last ninety days. While the headlines talk about 1.9% tech growth, those of us in the trenches are seeing a different reality: The floor has been hit. We are no longer in the "automation at all costs" era. We have entered the era of Human-Led Resilience. The Reality of 27-Second Breakouts In my day job in public safety communications, "uptime" isn't a KPI; it's a lif
The Autonomous Paradox In 2026, we’ve moved past simple chatbots. We are building Production-Grade RAG pipelines and autonomous agents that can plan, execute, and iterate. But as an architect, I’ve noticed a glaring hole in our "Agentic" future: Identity Sprawl. We are giving agents non-human identities (NHI) with "Full Admin" permissions just to ensure the RAG works smoothly. We are effectively
The math isn't complicated. It's just that nobody runs it until they get the bill. An AI agent handling a 10-turn workflow — reading files, calling tools, revising output — doesn't cost 10x a single query. Because transformer inference processes the entire context on every call, cost compounds with each additional turn. The tenth turn carries everything that preceded it: the original file reads, e
Every few years the industry rediscovers that programming languages are not religions. Then we immediately behave like they are religions. Someone posts a benchmark. Someone else says memory safety. Someone says developer experience. A distributed systems person appears from under a bridge and whispers “Erlang solved this in 1998.” A startup founder announces they are rewriting their CRUD app in R
Or: how we learned that “eventually” isn’t good enough when you’re bleeding file descriptors Or: how we learned that “eventually” isn’t good enough when you’re bleeding file descriptors Deterministic cleanup means knowing exactly when resources are freed — the difference between memory chaos and predictable system behavior in production environments. So our video transcoding service was… how d