Post

1/6 People often say LLM agents are “just another abstraction layer.” I think that confuses abstraction with automation. A useful tool is not automatically an abstraction.

Emmanuel Valverde Ramos

@emmanuelvalverde.dev

29 June

2/6 Abstraction is not just hiding complexity. It is building a model that lets us safely ignore details because the behaviour we depend on is stable enough to reason about.

Emmanuel Valverde Ramos

@emmanuelvalverde.dev

29 June

3/6 A good abstraction: - hides irrelevant details - preserves important semantics - exposes a clear contract - reduces knowledge consumers must carry - makes change safer

Emmanuel Valverde Ramos

@emmanuelvalverde.dev

4/6 Abstraction is not: - automation - delegation - orchestration - compression - a black box - “it usually works” Those things can be useful. But usefulness is not abstraction

29/6/26, 2:40 pm1 repost1 like

Emmanuel Valverde Ramos

@emmanuelvalverde.dev

29 June

5/6 If a tool forces you to understand its hidden quirks, model-specific behaviours, prompt sensitivities, or implementation details to use it safely, then it has not fully abstracted them away. That is closer to a leaky abstraction.

Emmanuel Valverde Ramos

@emmanuelvalverde.dev

29 June

6/6 LLM agents can be powerful. They can help explore, draft, generate, and automate. But calling them an abstraction layer too quickly hides the real cost: verification, review, ownership, and responsibility. Sometimes a tool is just a tool. emmanuelvalverderamos.substack.com/p/agents-are...

emmanuelvalverderamos.substack.comAgents are not a new abstraction layer. They are a new place to hide the billLLM agentic development can be useful, but calling it an abstraction layer is how teams rename verification cost as progress.