companies with ML problems with large scale data

Google
hktt

Go to company page Google

hktt
Mar 27 13 Comments

I am a SWE working in applied NLP and have some experience with multimodal data. I am looking for non-fang or pre ipo companies that deal with massive data like G scale or potentially equivalent like Discord. I am especially interested in multi-modality aspects.

Companies i have in mind:
Twitter
Linkedin
Pinterest (some interest)
Reddit
Snap (some interest)
Discord
Roblox #roblox (not sure about their ML problems?)

Any suggestions to the above list?

tc 350k

comments

Want to comment? LOG IN or SIGN UP
TOP 13 Comments
  • Congratulations! You're looking for pre IPO companies but all those that you listed are public. Dude ๐Ÿคฃ๐Ÿคฃ๐Ÿคฃ
    Mar 27 3
    • Oracle
      JSnowflake

      Go to company page Oracle

      JSnowflake
      I was about to say those are all bad examples, as they are public. Generally the posters requirements don't match up - most pre-IPO companies don't have massive scale data, or at least not proprietary sources of it. They usually have moderate scale data, shitty processes, and you are getting paid to come pick up the pieces of the shitty early stage employees who came before you, left a mess, and took most of the equity.
      Mar 27
    • Google
      hktt

      Go to company page Google

      hktt
      OP
      Ah i mean non fang or pre ipo companies. I mean their potential to deal data scale like at G
      Mar 27
  • New / Eng
    a420z

    New Eng

    a420z
    Twitterโ€™s cortex research team could be a good fit
    Mar 27 1
    • New
      outofname

      New

      outofname
      How is Search and Recommendations ?
      Mar 29
  • Amazon
    confuzd

    Go to company page Amazon

    confuzd
    Cruise, Waymo - not NLP but large scale data and very practical problem
    Mar 27 0
  • New / Eng
    a420z

    New Eng

    a420z
    Do you want to do modelling/research or infra?
    Mar 27 3
  • Google
    wUyR14

    Go to company page Google

    wUyR14
    I work on CV at G, and when I interviewed outside, I find most ML teams outside Google are just doing data engineering work. Unlike what we do at Google, they just plug-and-play existing model architectures instead of making any innovations on SOTA models.

    I would only recommend Meta or Nvidia. So far they are the only ones I know that do decent innovations in ML.
    Mar 27 0