@mattpocockuk: My hot take from chatting to @poteto is that we should use MORE abstractions in the AI age You can use them (combined w…
Summary
Matt Pocock argues that developers should use more abstractions in the AI age, since they constrain agents to good design decisions, improve token efficiency, and make unwinding bad abstractions cheaper when working with AI coding agents.
View Cached Full Text
Cached at: 10/03/26, 07:01 PM
My hot take from chatting to @poteto is that we should use MORE abstractions in the AI age
You can use them (combined with harsh lint rules) to reduce the design space available to the agent and constrain them only to good decisions.
Combined with the fact that high-leverage abstractions let you do more with less code - so, more token efficient.
Plus, unwinding the damage from a bad abstraction is much cheaper with agents.
This runs counter to a lot of folks thinking that agents just want to read the raw code. They can, but they’re not maximally efficient that way.
Be braver! Design abstractions.
Similar Articles
@mattpocockuk: In a phrase, "make bad code unrepresentable"
Matt Pocock argues that in the AI coding era, developers should embrace more abstractions combined with strict lint rules to constrain AI agents to making only good design decisions.
@mattpocockuk: Uncle Bob gets it
Robert C. Martin (Uncle Bob) shares his experience that using AI agents significantly boosts productivity, though 30-40% of time is spent tuning and hardening the output.
@rohit4verse: AI didn't make code cheap. it made bad code lethal. Matt Pocock: "Software fundamentals matter more than ever" AI in a …
Discusses how AI amplifies code quality, emphasizing that software fundamentals matter more than ever, and recommends five design patterns for building reliable AI agents.
@mattpocockuk: One reason AI sucks at design is because you're bad at telling it what you want Experienced designers have a thousand s…
Matt Pocock suggests that AI's poor design performance stems from users' inability to articulate design intent using designers' specialized shorthand, highlighting a skill gap in communicating with AI.
@mattpocockuk: Which codebase design techniques from the pre-AI era still help agents? Deep modules, adapters/seams etc. Writing the '…
The author is discussing which codebase design techniques from the pre-AI era are still useful for AI agents, as part of writing a course on codebase design.