What are the best ways to pick up Deep Learning skills as an engineer?

If you want to start implementing deep networks, you will need to learn tools like TensorFlow or Caffe or Theano or Keras. I would recommend TensorFlow (which is pretty new and is constantly improving with time. I haven't myself gotten around to learning it, but it is something I want to do in the near future). If you want to do deep learning on clusters, you will probably need to learn Spark or Hadoop. Getting a good handle on these tools is probably as important as learning about the math behind the models.