Hello friends,

A lot has changed since my last post. Previously, my trading process went like this: get an idea for a trading strategy, test it, deploy it and hopefully profit. However, this method of doing things is a little bit outdated and less sexy than what I’m doing now which is the following: have a computer program get an idea for a trading strategy, test it, deploy it and hopefully profit. I believe this is the future of trading.

How am I getting computer programs to generate trading ideas? I do this by using Deep Learning. In particular, I build neural networks in Python using Pytorch.

My inspiration to do this is Jim Simons, the founder of Renaissance Technologies, the most successful hedge fund of all time. However, they aren’t trading the Bitcoin market. That’s where your boy Shkreli comes in. I am using the same principles to trade that Jim Simons believes in, but applying them to a different market.

In particular, my neural network (form of AI) does not try to foresee much into the future, it can only tell me what trades to place today. It’s my job here to translate what I feel are the implications of what the AI tells me. I do that by applying years of my own years of experience.

Here are my thoughts on the current state of the market. I do not try to time the market or anything like that, so I can only state the obvious: we are by my metrics in a bear market and as I don’t try to predict market changes, it is my opinion we are probably going to be here for the near future. It could be a few years before BTC hits a new ATH. Since everyone enjoys making and reading about predictions, it is my personal belief that we will see a $10,000 or so BTC price before a $100,000.

Be wise,
YoungShkreli

PS: I actually do not possess a means of shorting btc through my exchange as I need $10M to do so and am not quite there :). So, while I have indicated short in the article, and while the AI thinks I should be short, I am not actually short. I am not capable of using leverage at all at the moment.