I especially liked the encouragement to talk to as many non-specialist colleagues as possible. The emphasis on continuous learning is good too.

I had some minor suggestions:

> Rely on strong fundamentals

You might want to add statistics and math here. However, you're so right to emphasise _fundamentals_! So often we have unidentified holes in our background.

Also, don't forget to add a suggestion in 'Always be learning': subscribe to DataTau!

Also, MOOCs such as Coursera and Kahn Academy for boning up on fundamentals. There are plenty of free resources. Then you can get your feet wet on Project Euler and, importantly, Kaggle. Some people swear by code boot-camps (but this seems to depend strongly on the boot-camp).

> Be decent at visualization

You've given some nice suggestions above. Why not some here, such as Shiny, ggplot2, Seaborn, mpld3?

> Document what you do

Another suggestion I would have is to try and choose tractable projects that you can stake ownership of. Perhaps try and make a cool infographic etc.

You can learn a huge amount from side projects and it can demonstrate you have initiative.

Finally, it's a little difficult to determine the scope of the article. Is it for people who already consider themselves Data Scientists? Or aspirants? If it's the latter, I'd also add that the job application process can be very revealing/informative. Join jobs networks like LinkedIn and Glassdoor and just look at what companies are asking for in terms of skills etc. If you feel ready enough, applying and hopefully interviewing can be instructive. Get in touch with other Data Scientists!