Hello,
After a month of using your product, I would like to suggest two possible enhancements:
-
In the context of research on a dataset, we often reach a point where two different paths or strategies appear. It would be extremely useful at that point to be able to duplicate the work so as to explore both strategies in parallel.
-
During the work on a Thread, certain code blocks become unnecessary or buggy and are automatically corrected. It would be great to have the ability to delete all code blocks that produce an error, in order to keep only the truly functional blocks. This would be very helpful, first of all for reviewing or exporting to a Jupyter notebook, and it would also allow for automatically re-running a thread to continue working with it (especially when returning to a thread after several hours or days).
Thank you for your work.
Sam