Cleaning data sucks so I spent a month automating it using agents and LLMs. If you'd like to use it or have feedback, let me know down below!

Everyone always tags @FarzaTV for feedback but they're sleeping on the feedback from @wordisbonz and @lentinidante my s5 toy for @_buildspace @_nightsweekends

Waitlist for ya

this looks cool, i think it would be helpful if there was a summary of what changed in the data from the the file which was uploaded. e.g. how many values were filled, outliers removed etc.

This is a great idea, it’s one of the next features i’ll implement in the LLM summary space.

i’d like it to queue up mutations with quick explanations so i know what it’s doing and can accept or deny the changes. less of a black box and more trust

I’m planning on adding a chat interface to allow smaller more precise mutations as well as broader data wrangling so that anyone can use it from hyper technical to no data science knowledge. Do you think that feature would meet your needs?

Great idea, will be super handy and helpful for data teams to let this app have a first pass at the raw data. And I agree with one of the comment here - that as an engineer or scientist. I would like to know what changed in my data and why (which LLMs should be able to do fairly easily).

I used your Local LLM tutorial tweets when I first started this project, this means so much coming from you!

Keen to try this out!

Keen to hear your feedback!