Large Changes
Heads up... You’re accessing parts of this content for free, with some sections shown as text.
Heads up... You’re accessing parts of this content for free, with some sections shown as text.
Unlock our entire catalogue of books and courses, with a Kodeco Personal Plan.
Unlock now
So far we’ve been really quite precise about what we’d like our different AI assistants to implement—carefully limiting what we’re asking to ensure that it doesn’t overstep its bounds. But we’ve got a detailed spec, and continuing in this manner will take quite a long time to complete Yesterday’s Weather.
We’ve got a complete and detailed spec for this project, so we can actually ask AI for an implementation plan, approve it, and then set it off to implement it.
This generalises to the following development process:
- Develop, and refine a spec with a chat-based LLM.
- Review and approve the spec, checking high-risk areas carefully.
- Ask an AI agent to write an implementation plan.
- Review the plan, if necessary refine it, or cut it down.
- Ask the AI agent to implement the spec according to the agreed plan.
- Perform a code review, and iterate as required.
We worked on (1) and (2) in the first lesson, and we’ll look at (3) to (5) in the next video.