JumpyTaco

Boss asking something shady

Boss is asking me to find selective samples in the data to where model performs good and then do a k-fold on the sub sample.

He’s saying that due to computational intensity we can’t run model on entire test set.

What would you do?

19mo ago

Jobs

One interview, 1000+ job opportunities

Take a 10-min AI interview to qualify for numerous real jobs auto-matched to your profile 🔑

+322 new users this month

DancingHamster

Stealth19mo

Lol happens everywhere bro. I know of a team where they were buying a sophisticated model using LLMs and a fucking logistic regression model was able to provide VERY similar results. Figure out what they had to do to show that their model was worth the effort :)