James Stevenson 
Bio:
I am a second-year engineering PhD student studying chemical engineering. But back in high school, in Manhattan, I thought I wanted to be a writer. There aren't many engineers in New York City, and my high school was designed to make writers and lawyers, so I didn't know what an engineer was until junior year. I would have told you that he was the man in the funny hat that drives a train. I wanted to invent things, but I thought you had to support that with a real job. Apparently it is a real job.
I studied undergraduate chemical engineering at a tiny school in lower Manhattan called Cooper Union. The place had no money; the laboratories were a step down from high school. The older students compared it to Tony Stark trying to build the Iron Man suit in a cave using a box of scraps. But their computers worked. One of the neat things about computers is that they are all equal - any computer can simulate any other (and you can mathematically prove this). So when I started doing research, I used computers instead of a lab.
An undergraduate degree was enough to show me that I knew nothing. I wanted to get to the cutting edge of the field, so I decided to seek a PhD.
In my spare time I bike, form eccentric business schemes, and argue about philosophy on the internet.
Research:
My work is developing algorithms. An algorithm is a way of breaking a big problem down into known parts. The known parts are things like the multiplication table: you just memorize it, and then you can multiply any small numbers. But for big numbers, like 322 x 865, you don't have space in your head to memorize all of them, so you break the problem down into pieces (2 x 5 = 10...) that you can do.
The algorithm I'm working on now is kind of like the multiplication table, but bigger. The idea is to make a computer remember all of the different ways that two water molecules can interact with each other, just like remembering how each two numbers interact (6 x 4 = 24, 9 x 10 = 90, etc). While a multiplication table has two dimensions, the table of water interactions has six dimensions: three distances and three angles.
It’s challenging to fit this six-dimensional model into even a computer’s memory. But the advantage is, remembering an answer that you already know is usually faster than working it out again. My water model should allow people to do accurate calculations about water molecules faster than ever before, which is useful when studying water-based things like ice, steam, and human cells.
The process which I most want to be able to model is the trapping of carbon dioxide in ice crystals (hydrates) and the liberation of methane. Existing water models are insufficiently fast and accurate to study this process, which could perhaps be used to extract vast quantities of natural gas without causing climate change.