AI agents will not fit neatly into fixed APIs or index-everything data platforms. They need adaptive data infrastructure that can interpret intent, compose datasets on demand, and optimize around real usage.

AI agents will not scale on rigid schemas and pre-built dashboards. They need composable data systems that expose small, well-scoped datasets, rich metadata, and interfaces agents can assemble on demand.

A perspective on why composability keeps reappearing across computing, from Unix pipes to blockchain protocols and AI agents, and what it means for the future of reliable data systems.