Subscribe to DSC Newsletter

Participate, and see results at

For a geographic breakdown of answers per city (all over the world), see

Views: 3596

Replies to This Discussion

In Los Angeles, difficult to find senior-level jobs; relatively easy for more junior people.

Big Data is very hot now and seems to be required at all levels.

Los Angeles is not a great place for analytic people. But if you can convince an employer that your analytic talent will boost revenue or reduce losses or fraud, you will be hired quickly and have little to no competition.

My experience in Germany is, that if you have a degree in a somewhat different field, and you don't have experience with the one special tool they point out in their job offer, your not even contacted from the hiring staff, no chance...

I used to develop analytic algorithms and models by myself not using a distinct tool but programming it myself in various programming languages for my research position I held in machine learning/statistical learning theory/computational neuroscience and I have lots of experience and in depth knowledge. I am still looking for a better fitting job, but lost quite a bit hope in getting one in Munich or even in Germany

Since this is a relatively new field, the supply of qualified quantitative workers is low and the cost high as companies seek people with specific degrees, perhaps advanced degrees with specific experience. This is just an HR check-mark with no correlation with reality. Perhaps HR should hire a quant to help them figure this out!

I think this will change significantly as the demand/supply for 3GL developers did in the 1980's- 1990's. In the 1980's you had to be a computer science major to develop. By the end of the 1990's, demand so outpaced supply that developers had every background imaginable.

You can see this in Michael O’Connell's (of Tibco) article in Forbes..

"Generally, college graduates with some propensity to learn new tools quickly tend to do better as corporate data scientists, than do lifelong academics, he says. A PhD statistician, for example, may not necessarily be the best candidate to be a data scientist in a business context, as they tend to work on small and arcane data sets, they may not have proper empathy for business problems, and may have been trained on archaic software, O’Connell says."

Ironically, I don't think you need any diploma to be a data scientist. You can acquire the knowledge (SQL, Hadoop, Python, machine learning, statistics, data processing, R) for free on the web. What you need most is intuition, good judgment, vision, be able to guess what the client wants to achieve, some craftsmanship, rule of thumbs, experience with various types of large data sets - things that you don't learn at school.

Amy, I am completely in agreement with you. I have done enough jobs where my schooling had little to no impact and it was what I was willing to do on the side that was useful. I wonder if this situation is yet another nail in the coffin of schooling as we know it.

Spot on, Amy. The intuitive ability to choose features and properly evaluate training and test results is the skill ... (i.e....understanding over-fitting, local and global minima, etc..)

IMHO it's as much an art as a science ... we should focus on apprenticeships instead of an academic credential for the mathematically gifted young ... to develop really effective quant workers ;-)

Hi Amy, I'm fairly new to the field and am curious about how Python fits in. Is that what is commonly used for data manipulation? Is there an advantage to knowing Python over another scripting language such as Ruby?

Michael O’Connell's quote was spot on.

I was a litle frustrated by Vizu's "Further Analysis" tab.  It would not give any drill down of the results.  It would be interesting to see the results broken down by age groups, gender, etc.


In the Bay Area, sounds it's difficult to find a job in you are in San Francisco, not easy if you live in the East Bay, and easy if you are in the Peninsula / San Jose area -- where all great "data science intensive" companies are located (Google, Facebook, LinkedIn etc.), as well as Palo Alto / Stanford University, the birth place of analytics engineering and big business data.

If you are mobile, there is a huge flood of opportunities, especially for experienced data scientists.  But there is a big challenge/gap right now for many people.  I've been on both sides of the hiring decision multiple times and built two different teams over the past several years. 

There truly is a strong need for the ubermensch data scientist, ala the Forbes article, but there is a short list of people who meet those qualifications.  Hiring managers who do not have a good understanding of how to balance a diverse team to get all of those skills in the right combination (read: most hiring managers), cannot give clear instructions to the recruiters they often have to rely on to find candidates.  Recruiters are far from understanding what experience is going to provide the right skills and capabilities and so can only judge resumes by the key words they have been given or researched.  I frequently run into situations where I am contacted about a position, I know a strong candidate who is well suited, but is subsequently screened by the recruiter because he or she did not have the right degree or the right stats software package listed, etc.  It is ridiculous.  It can also be frustrating for those that don't have someone who can call up the hiring manager and say, "You need to take a second look."

When I am hiring, and most experienced, analytical hiring managers I know do this as well, I do not rely on recruiters alone, but use personal channels, LinkedIn and any other resources available, like AnalyticBridge.  When I am looking for a job, I reach out to my network and always have multiple offers in a couple weeks.  It really is all about who you know, and frankly, if I am looking for an ubermensch, I know they will have a network and will come recommended.

Just my $0.02.


Follow Us

On Data Science Central

On DataViz

On Hadoop

© 2016 is a subsidiary and dedicated channel of Data Science Central LLC   Powered by

Badges  |  Report an Issue  |  Terms of Service