philippsinger.com
Senior Principal Data Scientist @h2oai
PhD in CS
Top ranked Kaggle Grandmaster (Highest #1)
All views are my own.
https://kaggle.com/philippsinger
21 posts
195 followers
301 following
Getting Started
Conversation Starter
comment in response to
post
the agent is literally mostly just for finding relevant files, which is super important, the main composer/coding agent is basically the same
comment in response to
post
but think about all the concerned users sending their data to a 3rd party app
comment in response to
post
He didnt scrape anything. He called an openly available endpoint.
comment in response to
post
Ok, lets say the vocal majority or rather "quite a lot". Just need to read the comments regarding the privacy discussions the last few days.
comment in response to
post
But everyone thinks they know everything 😁
comment in response to
post
yes, and you can get this data via API, there is no need to scrape
comment in response to
post
Adguard was first to run :)
comment in response to
post
Yeah, that one will be setup
comment in response to
post
Just run it separately via uvicorn. Mostly with docker containers. But if you want to run it in eg vercel you can do so directly: vercel.com/templates/ne...
comment in response to
post
As long as the author knows, it is not truly anonymous though. So the benefit is minimal in terms of that concern.
comment in response to
post
please no, why would they
comment in response to
post
Using wandb itself is a misery. I really dont get why so many people like it while there are much better alternaitves.