4 ms·
Assuming you actually can hire the best people, and somehow do so in an affordable manner when half the planet applies, that's a great strategy. But I'd expect
by nitwit005 3y ago
Assuming you actually can hire the best people, and somehow do so in an affordable manner when half the planet applies, that's a great strategy.
But I'd expect a very low number of firms to succeed at that.
- hiAndrewQuinn 3y agoNow see that's an interesting problem space to be in. Keen eye. To me it seems like a really exciting place to apply recent innovations in LLMs. If LLMs can chew through thousand page legal binders like toilet paper, there's no reason they can't chew through a thousand 1-page resumes and spit out "These are the ten most promising ones based off of our statistical analysis." Firms can specialize in the production, hosting and fine tuning of these LLMs, and even play both sides of the market by allowing candidates to see how good their resume looks for a given job description. I think this is much likely to become a lot more common in the latter half of the 2020s.
- valzam 3y agoAnd then you try to look at the top 3 candidates and realise the LLM hallucianted them all
- __loam 3y agoYes lets make applying to jobs even more of an algorithmic hellscape.
- bigstrat2003 3y agoWhether or not it is pleasant for the applicants, it'll happen if it provides a benefit for the companies. They don't care about the applicant experience because they have no incentive to.
- Draiken 3y agoThis is the dystopian future that is likely already happening. Slowly but surely we will lose control over our own labor. As the name implies, we're just human resources.
- marcinzm 3y agoResumes are candidate controlled which inherently makes them useless once social rules on lying too much break down.
- notpachet 3y agoThe flipside of this is that there will be an asymmetric advantage available to firms that are capable of finding excellent candidates that fall into the ML blind spots.