# Who is using ultrathink? Whats your strategy? — retained record

> Retained reddit record from reddit:ClaudeCode.

Canonical: https://highsignal.app/data/records/1ed36800b4ed5dc7

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:ClaudeCode)  published 2026-08-29 18:01 ingested 2026-08-30 02:39

[original source ↗](https://www.reddit.com/r/ClaudeCode/comments/1w1tdbw/who_is_using_ultrathink_whats_your_strategy/)

## retained text



```
I've searched and found pretty much zero discussion on this feature. Which is weird because it seems like it should hypothetically be useful if used properly. The ability to selectively maximize the effort/reasoning on a single turn, when applied to a complex problem, seems useful​

I'm exploring ways to use it, but claude often takes multiple turns between responses, so I have a hard time knowing for sure that the "ultrathink" lands on the turn where it matters 

The strategy I've been using is: spend a session crafting a long prompt with full context of the problem. Then take that prompt into new session where I instruct claude to ultrathink, with the specific goal of creating a spec/implementation plan for the solution to the problem

This strategy feels potentially wasteful and overly token intensive because I end up spending 3 sessions. So I'd love to hear what anyone else in the community is doing to make use of the ultrathink feature. TIA
```



document fetched 2026-08-30 00:36 · canonical https://reddit.com/r/ClaudeCode/comments/1w1tdbw/who_is_using_ultrathink_whats_your_strategy
