SpaceBook started off as just a "facebook" pun in my head, but when I mentioned it to my group they were immediately on board. Seeing as none of my other group members had experience or confidence in full stack software design, I took charge of building the majority of the architecture for the software. I ended up learning a lot from my mistakes in both designing the database, choosing information flow, and client side caching. I gained a lot of experience with project management when it comes to full stack applications.

I figured if I was going to make a game, it had to be a difficult one since I'm a "pretty skilled" developer. I may or may not have severely underestimated the complexity of a game like Texas Holdem. Oh well. There were a trillion edge cases I didn't plan for, and I had to refactor the code several times. I learned a lot about the importance of planning and clean, maintainable code. I also got to try some cool animations and transitions with CSS and JavaScript, although my current animation skills now put the ones in this project to shame(or so I think).

We genuinely had no idea what we were doing. We knew we needed to detect automated accounts, that's it. There was a fuzzy time period for almost 2 weeks where we were just doing research trying to figure out how this would be done. I had to research alot into how to use machine learning, I came to the conclusion on a clustering algorithm. Even then, the model didn't work as planned due to poor data processing/transformation. I implemented some of the complicated techniques like applying a logarithm before a our standard deviation normalization. The model ended up working pretty well, we couldn't find many mistakes the model made in our analysis.