Data & Management
Antifragility in technological systems. What Fable 5 demonstrated
Most AI stacks built in 2024 and 2025 were fragile. They didn't know it because the pressure had not yet arrived.
The Logbook
Exploration of ideas, essays, and developing notes on technology, sustainability, and design.
Data & Management
The Fable 5 Episode and What Changed in My Stack
A well-designed stack does not collapse when a provider fails. What Fable 5 demonstrated was not a failure of Anthropic — it was an architectural failure in all stacks that depended on a single point of jurisdictional failure.
Data & Management
Why I Stopped Being Loyal to a Single AI Model
There is a difference between using AI and working with AI. The former is loyalty by default. The latter is architecture by choice.
Models and Routing
How I Distribute Work Among Eleven AI Models Without Relying on Any
Operating with a single AI model is convenient until that model fails, censors a task, or charges you three times for reasoning that another does just as well. This resource documents the orchestration architecture I use in production from the Ambalá-Calambeo corridor: eleven models with fixed roles, a failover matrix, and prompting rules that prevent the AI from inventing business rules.
Do you want to dive deeper into a topic?
If any of these notes resonated with you or you'd like to collaborate on an idea, let's talk.