AI Architecture Updates: May 13, 2026

1. Unmesh Joshi reframes code as conceptual model in the LLM era

Unmesh Joshi, writing on martinfowler.com, argues that code serves two purposes: instructions for a machine and a conceptual model of the problem domain. As LLMs make code generation cheaper, the mechanical act of writing instructions becomes less central, while well-structured code with stable abstractions becomes more vital because it acts as context and guardrails for LLMs producing reliable output. Source