Thanks to LLMs, I think making one "dataset" match one file could be interesting. Easy to run on isolation, debug, iterate, ...
Thanks to LLMs, I think making one "dataset" match one file could be interesting. Easy to run on isolation, debug, iterate, ...