Yeah, the schema driven approach does really lend it to the world of LLMs. We've found that with minimal guidance LLMs can generate schemas, validate them and iterate if necessary.
What's great about this is that once you've got a valid schema you can then guarantee the correctness of the code downstream as well as it being very testable. This text-based abstraction makes it possible to generate a complete production-ready backend with AI rather than trying to get a LLM to write complete applications from scratch.
A weekend project I created to monitor activity of an app we just launched. Chirpy generates audio melody from your realtime analytics data. You an quite literally listen to your users!
Chirpy is still in the early stages (especially the tracking modules) but really interested to see what you guys think!
What's great about this is that once you've got a valid schema you can then guarantee the correctness of the code downstream as well as it being very testable. This text-based abstraction makes it possible to generate a complete production-ready backend with AI rather than trying to get a LLM to write complete applications from scratch.