# Why LLMs Struggle to Maintain Mental Models in Software Engineering

DevFeed: [Why LLMs Struggle to Maintain Mental Models in Software Engineering](<https://devfeed.tech/articles/why-llms-can-t-really-build-software-13566.md>)

Original publisher: [Read original article](<https://zed.dev/blog/why-llms-cant-build-software>)

Author: Conrad Irwin

Published: 2025-08-14T00:00:00Z

Content type: opinion

Language: en

Sources: [Zed Industries - Blog](<https://devfeed.tech/sources/zed-industries-blog.md>)

Topics: [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Software Engineering](<https://devfeed.tech/topics/software-engineering.md>), [Code](<https://devfeed.tech/topics/code.md>), [context window](<https://devfeed.tech/topics/context-window.md>)

Tags: [code](<https://devfeed.tech/tags/code.md>), [context](<https://devfeed.tech/tags/context.md>), [context-window](<https://devfeed.tech/tags/context-window.md>), [debugger](<https://devfeed.tech/tags/debugger.md>), [llms](<https://devfeed.tech/tags/llms.md>), [software-engineering](<https://devfeed.tech/tags/software-engineering.md>), [tests](<https://devfeed.tech/tags/tests.md>)

## AI overview

The article argues that effective software engineering depends on building and maintaining clear mental models of requirements and code, not merely writing code. It says current LLMs can generate code and perform supporting tasks, but often become confused when tests fail or when they must decide whether to change the code or the tests. The author suggests that improving this may require models with better ways to manage context and reason about problems.

## Source excerpt

Writing code is only one part of effective software engineering.