Job Skills shows no mention of R. I find that surprising. I believe there are still a lot of Statisticians/Applied Math folks still untapped in the startup community.
The article lists smaller startups. Data scientists at these shops need to contribute to production systems, which means Python and C/C++ more than R or Matlab.
I think that the lack of mention of R just points out that most applied data science oriented startups are for the large part run by computer scientists / software engineers rather than statisticians or applied mathematicians.
R is heavily fragmented and requires at least some willingness to read papers, identify the right package for the right task, and in many cases interpret non-standardized output.
Python on the other hand has 2 consolidated modules - numpy and scipy - that for most basic data science tasks are good enough and easy enough for someone not trained in statistics to understand.
Indeed, it is a bit weird. Note: This was no exhaustive search for job skills, I think that deserves a single post, with well... proper data and fancy graphs (preferably made in R).
I am sure there are new data science startups looking for R-magicians. No doubt some of the best use it, as seen in this slideshow by Xavier Conort (10 R packages to win Kaggle competitions): http://fr.slideshare.net/DataRobot/final-10-r-xc-36610234
On another point I understand the startup community is dominated by hackers with the standard programming tool box like C, Python, Java, et al. So I believe there is still opportunity.