Posts tagged ai-systems
Before You Add an Agent, Build the Harness
The hard part of using agents in production is not getting a model to produce an answer. It is deciding what authority the system gives to that answer.
Your Agents Are Already in Production. Your Platform Isn't.
Most teams don't have an agent platform problem. They have an accountability decision they haven't made.
Constraint-Driven Thinking: Why Most AI Outputs Are Useless
Without constraints, systems don't make decisions. They generate convincing options.
You're optimizing the wrong thing in AI coding systems
Most teams focus on outputs. The real problem is deeper: you're asking a probabilistic system to behave deterministically — and that requires a different kind of engineering.
El agente hizo lo que le pediste, y ese fue el problema
Por qué la autoridad es el primer control que necesita un agente de ops — y el último que se configura