[maerch]

> A repeated trend is that Claude Code only gets 70-80% of the way, which is fine and something I wish was emphasized more by people pushing agents.

Recently, I realized that this applies not only to the first 70–80% of a project but sometimes also to the final 70-80%.

I couldn’t make progress with Claude on a major refactoring from scratch, so I started implementing it myself. Once I had shaped the idea clearly enough but in a very early state, I handed it back to Claude to finish and it worked flawlessly, down to the last CHANGELOG entry, without any further input from me.

I saw this as a form of extensive guardrails or prompting-by-example.

[golergka]

That’s why I like using it and get more fulfilment from coding than before: I do the fun parts. AI does the mundane.

[bavell]

I need to try this - started using Claude code a few days ago and have been struggling to get good implementations with some high-complexity refactors. It keeps over engineering and creating more problems than it solves. It's getting close though, and I think your approach would work very well for this scenario!

[LeafItAlone]

The best way I’ve found to interact with it is to treat it like an overly eager junior developer who just read books like Gang of Four and feels the need to prove their worth as a senior. Explain that simplicity matters, or you have an existing pattern to follow, or even more specific.

As I’ve worked with a number of people like what I’ve described above, the way I’ve worked with them has helped me get better results from LLMs for coding. The difference is that you can help a junior grow over time. LLMs forget after that context (Claude.md helps, but not perfect).

[theshrike79]

Claude has a tendency to reinvent the wheel heavily.

It'll create a massive bespoke class to do something that is already in the stdlib.

But if there's a pattern of already using stdlib functions, it can copy that easily.