What’s your best LLM workflow for planning vs execution?
When building things like apps, websites, designs, PDFs, or learning docs:
- Which model do you use to plan?
- Which model do you use to execute?
- Do you split planning and coding/building across different models?
- For coding, do you use Codex, Claude Code, or something else?
- Does your workflow change by task type?
I feel like planning is the most important part, so I want to know which model is strongest at planning, and which is strongest at doing.
Would love to see your actual workflow step by step, not just model names.
[link] [comments]