About 6 months ago, I threw together a quick & dirty (unregistered) package that I called DataMimic for generating synthetic data, based on a real dataset. It was okay for what I needed it for at the time, but it’s definitely low quality, and now I would like to revisit it and turn it into a proper package. I have had success with spec-driven development and Claude, so I want to take that approach here, but I want to leverage the incredible human expertise in this community to get feedback on the spec:
I’m particularly interested in:
The Lux over Flux logic
The type system design
The proposed public API
And then anything else folks think should be changed/improved/etc. Thanks!!