DS - Glorified SQL Monkey

Nov 27, 2019 17 Comments

I hear people complaining that at some companies Data scientists are not doing ML work. Instead they're building reports and doing ad-hoc analytics.
Which companies are those?

comments

Want to comment? LOG IN or SIGN UP
TOP 17 Comments
  • New fsgo55
    90% of all data science jobs in banking are just putting reports together with SQL or pandas. 9% of data science jobs in banking use linear regression and maybe 1% use something else
    Nov 27, 2019 5
    • T-Mobile / IT ensemble
      @Demos0923B, you've seen that at Microsoft?
      Nov 27, 2019
    • Microsoft / Data
      Demos0923B

      Microsoft Data

      PRE
      General Dynamics Information Technology
      Demos0923Bmore
      I work services, so I work with our customers. No idea about internal, but I have seen on a couple customers and before I was at Microsoft I saw it all over the place. Generally finance people are Excel magicians, and attempts at brining them out of the 90's often fail.
      Nov 27, 2019
  • Oracle / IT
    l00tb0x

    Oracle IT

    PRE
    Ericsson, National Instruments
    l00tb0xmore
    Most of them.
    Nov 27, 2019 0
  • Snapchat
    rThn06

    Snapchat

    PRE
    Facebook
    rThn06more
    Facebook
    Nov 27, 2019 0
  • WeWork / Mgmt
    WeProfits

    WeWork Mgmt

    BIO
    CEO of the We Company.
    WeProfitsmore
    T-Mobile and Wework
    Nov 27, 2019 2
    • T-Mobile / IT ensemble
      😁
      Nov 27, 2019
    • T-Mobile jensen1
      Hope you are still floating bro
      Nov 27, 2019
  • New / R&D OK👻mer
    Any req that doesn't say PhD preferred. What you're really after is applied or research scientist roles if your goal is to do modeling and use a bit of math.
    Nov 27, 2019 2
    • Microsoft / Data
      Demos0923B

      Microsoft Data

      PRE
      General Dynamics Information Technology
      Demos0923Bmore
      This is probably a good indicator. Also, not sure if it actually helps (haven't looked), but maybe apply in the sciences or healthcare fields where they really need pure Data Science to solve a problem (e.g. genetic sequencing, etc.). These also tend to have deep pockets too.
      Nov 27, 2019
    • New / R&D OK👻mer
      I was going to suggest something similar as well. Products that are based on the natural sciences or physics will tend to always require interesting models.
      Nov 27, 2019
  • Sony / Data
    iXgsmg

    Sony Data

    PRE
    King
    iXgsmgmore
    Anything product heavy is usually SQL and reporting heavy, so practically any application, game or web service
    Nov 27, 2019 0
  • IEEE AmInuts
    Coming from a different background (quant Finance) I found the terms ML and DS to be somewhat misleading in the tech world. ML usually refers to ML Engineers (notable exceptions include Microsoft and Google Research) who can build to scale. DS seems to be encompassing just about everything.

    I interviewed for Staff Data scientist at LI and the stat questions were certainly Ph.D / Post Doc level questions. I explicitly mentioned I do not know Python, SQL or R.

    At LI, things like variational inference methods, Stochastic Multi Arm bandit and Contextual Bandit are considered well under the DS umbrella. However from what I heard from friends at Amazon, they are considered to be under the ML umbrella. And DS seems to be relegated to SQL querying and such.

    Would love to hear about this from people at other companies, especially Apple and Google.
    Nov 27, 2019 0
  • Oracle P1m77
    Varies wildly by team. Easiest way to find out is look at LinkedIn profiles and read what people do. Not a guarantee, but it's as close as you'll get to finding out ahead of time.

    One correction to OP, "advanced analytics" teams usually have both ML and statistics and SQL people. The ML and statistics is what makes them "advanced," but not everyone on the teams does everything.
    Nov 27, 2019 0

Salary
Comparison

    Real time salary information from verified employees