Related Posts
Hi,
I need some consulting.
I am a fullstack developer having 8+ years of experience,I was enjoying my work but now I think I hate coding.Currently I have changed my job.now I Am just fixing the bugs.nothing new.I started hating coding and also I can't take up stress.Now I have decided to change my domain.But not sure which domain to pick and don't want to go from start.Can anyone suggest something which domain to pick up.as I was thinking to go into techno-functional consultant or BA.can someone suggest
More Posts
Bain & Company What consulting firm would you pick if given offers from every single MBB/Tier 2/Big 4 firm and WHY?
PwC Deloitte Accenture EY EY-Parthenon Strategy& L.E.K. Consulting Alvarez & Marsal Alvarez & Marsal Oliver Wyman McKinsey & Company Bain & Company Boston Consulting Group
Additional Posts in Data Analytics & Data Science
New to Fishbowl?
unlock all discussions on Fishbowl.



The title data scientist can be super broad just like software engineer. In most cases, the minimum skills necessary include manipulating relational data with some form of SQL, working with big data (out of memory), and writing programs that manipulate data and do statistical operations at a no-trivial scale. I'd say you pretty much have to know some python and it's common data tools, but there are always exceptions. You certainly need to understand some application of statistics very well either theoretically or practically. This could be A/B testing, forecasting, optimization, or ML (itself with many subdomains to focus on). I tend to take the word scientist in the title seriously. This means a proven track record of turning raw, novel data into novel insights or broadly predictive models.
I wouldn't really expect someone to check all those boxes before you even start training them. But what I would say is that you should see the prospect of them being able to check all those boxes (and hopefully more) in the near future. You should see a lot of internal motivation, because as you say teaching is hard and none of us really have much time for it. I don't even care if someone knows SQL at all, or knows how to build an ML model on big data in Spark, or knows the relative merits of every statistical test. But, you should see the prospect that they could figure that kind of stuff out without a huge amount of hand holding. They should have very good data sense and impress you with what they can get done with the skills they have already. It would be a good idea to keep tabs on them as they take a bootcamp, as you are thinking, and observe how much they get out of it. It's probably a good minimum threshold before investing much of your own time.