We built a small playground for a tool that dynamically segments unstructured text data (support tickets, chats, or call notes, ...) with automatic dimension and segment discovery over time.
We started this while working on analytics for AI workflows and realised there were some interesting engineering challenges behind it like dynamic dimension discovery and merging and scalability of the data processing pipelines.
Try it here: https://www.correl8.ai/playground
We'd love some feedback on the core value and if you see anything to improve or add ?
Thanks! Romain Zimmer