ScrexyScroo

LLMs at best are a good auto-complete. I have been trying to use LLMs to code since GPT 3 came out. While Claude 3.7 has made some progress still it can't generate an app into existence, although it's great for boilerplate or hinting to which direction I should try to find the documentation.

At my workplace which deals with millions of customers. What we have is essentially a segmented clone of prod. It's a 1:1 copy of prod with real data flowing through.

We use feature flags to enable-disable features. This way when our devs ship code to prod - it first lands in this segmented clone.

Then incrementally changes are propagated from this segmented area into actual prod-prod.