My name is Tristan. I’m currently unemployed, and I’m using my free time to get back into programming. I used to use Linux a lot, so I’ve got some experience in Bash, as well as a tiny bit of Python. In high school I took a Java class that didn’t help me much, because the “teacher” was learning with us, and I ended up helping him teach the class. I don’t remember any Java at this point. I briefly did web design as well, but only touched HTML and CSS.
I also do a bit of crypto trading, and I’ve gotten fairly decent at it, so I wanted to try my hand at writing an automated trading bot. Being unsatisfied with programming a strict strategy, I’ve decided I’m going to learn as much as I can about neural networks and try to implement my own intelligent trading bot. On that subject, I’ve noticed all of the bots others have created have been “future price predictors”. Not only does this sound like a not-very-efficient way of designing a bot, I don’t think it will ever work properly. This is why I have come up with my own model idea, in which the model doesn’t even think about future prices. Instead, it will operate sort of like a machine vision model, and will learn to recognize shapes in the charts, and train itself based on profit over time instead of accuracy of a price prediction, much the same way as a human would.
Another project idea I have is for a complex cell evolution game, kind of like Cell Lab and the cell stage of Spore. These games are always cool for a bit, but very disappointingly simple (especially Spore). They don’t truly do what I’d consider cellular evolution. So someday I’d like to design a cellular evolution simulator/game.