Is ML/AI hype over?

Oracle
safradogz

Go to company page Oracle

safradogz
May 29, 2020 15 Comments

Watching layoffs at Uber/Lyft/Airbnb/etc, most of RIFs are from either DS or ML infra teams.
Looks like it is not resistant to recession.
Thoughts?

145 PARTICIPANTS SELECT ONLY ONE ANSWER
VOTE VIEW RESULT

comments

Want to comment? LOG IN or SIGN UP
TOP 15 Comments
  • NetApp
    KrishnaTC

    Go to company page NetApp

    KrishnaTC
    You got a logical fallacy there: exactly how is it different from asking "is software hype over?" after some big companies fire SWE (as well as other employees)?
    May 29, 2020 4
    • Lyft
      tr74gy

      Go to company page Lyft

      tr74gy
      Not true. DS team was not more affected at Lyft.
      May 29, 2020
    • NetApp
      KrishnaTC

      Go to company page NetApp

      KrishnaTC
      Here is a likely possibility that can reverse your perspective:

      The companies laying off ML teams may have achieved a certain point when the ML infra is good enough to generalize over most problems and allow the company to print money for decades.

      Theoretically, you can imagine that Google could transfer 95% of all ML employees working on Ads, without critical damage.

      Another possibility, some ML teams (besides infra), applied research ones, work on long term features. This is definitely "nice to have", because not all companies need this kind of innovation. This is true especially when they have a working product and an unclear vision, focusing on survival.
      May 29, 2020
  • Well thats absurd. Layoffs are always based on cost saving and business importance. Its always crucial that you have devs to support but thinking that ML is hype just because of the layoffs is stupid. ML is always valuable and you cant deny that. When times are bad you try to survive and I dont think survival will be the mindset forever.
    May 29, 2020 5
    • Google
      Prometh3us

      Go to company page Google

      Prometh3us
      My take is that ML will keep growing, and people with legit ML skills will not have much trouble finding jobs. Many of the current ML practitioners are folks who have functioning knowledge of ML along with reasonable SWE skills, but not great at either. Those are likely to be the first cuts, especially as ML gets democratized and engineers who can “do ML” will be most undergraduate degree holders who are available for cheaper.
      May 29, 2020
    • Google
      humble00

      Go to company page Google

      humble00
      This ^ those w/ fancy ML titles but w/o solid SWE skills will be cut.
      May 30, 2020
  • Google / Eng
    WVXD02

    Go to company page Google Eng

    WVXD02
    My professor used to say the AI hype train comes and goes. If it dies off it will be back.
    May 29, 2020 2
  • Google
    Prometh3us

    Go to company page Google

    Prometh3us
    You may be conflating DS and ML and AI.
    May 29, 2020 0