Can pseudocode cut token costs before refactoring code?
A messy project could be by first asking Codex to turn the existing code into . is easier for a person to read and improve before changing the real code. can create extra cleanup work, and that cleanup may use many tokens.
One possible workflow is to keep notes while building with AI, improve those notes from time to time, and then use them to guide code changes. The open question is whether this would save tokens and help with large , or whether maintaining the extra notes would create more work and cost.
Key points
- Turn messy code into before doing a larger refactor.
- Use as a simpler map of how a project works.
- This may reduce tokens if the model can read the map instead of the whole codebase every time.
- The approach can backfire if the becomes outdated or misleading.