We’re talking about a machine that’s capable of a billion billion calculations per second, or one exaflop. Every person on earth would have to do a calculation every second of every day for over four years to match that. Researchers could use the power to run massively complex simulations that can help advance fields like climate science, genomics, renewable energy, and artificial intelligence.

What if our brains could function at exaflop speeds? They will and they can. Waiting for Sapien 2.0

Learning AI

Here are some great resources to get you started:

  1. Practical Deep Learning for coders ( you must have at least 1 year of coding to follow the examples )
  2. Practical Deep Learning Part II
  3. MIT’s deep learning course
  4. The best data science blog from one of my favorite people Rachel Thomas
  5. The Montreal Institute for Learning Algorithms – MILA

I will add to this post as I discover new resources.


For those of you thinking about tackling AI or wanting to learn “how to AI” here are some myths about getting into AI:

  • You need a deep learning PhD
  • You need “big data”
  • Deep learning replaces domain experts
  • You need lots of expensive computing power
  • Deep Learning only works for very limited problems
  • AI experts are scarce and hard to find

That’s all bull. Dive into AI and make something happen!


It is obvious to me. . .


that “killer robots from hell” is a real imaginable thing. As robots become ubiquitous along with AI, it’s inevitable these two technologies will merge and enhance one another. I find that there are so many believable scenarios that tell the story of The Matrix over and over and over again.