I hosted a small workshop at KTH (where I'm currently pursuing a master in Machine Learning), presenting Python's capabilities as a tool for data-processing (or "data science" if you prefere fancy terms!) and machine learning.
You can view the slides on Speaker Deck (or embedded below), and I highly recommend having a look at the accompanying IPython notebook. (all sources are available, including the notebook, are available on this repo)
Note: Please disregard the slides where I encourage people to use Python 2.7 over Python 3. That's a controversial stance and one out of place for such a presentation, sorry!