Fun Pokemon Project
Hi! I’ve been currently working on a project that takes replays of matches (about 56,000 games) from a Pokemon battle simulator (Pokemon showdown) and parses their downloaded version that is in a json format to create an analytics dashboard about Pokemon team usage rates ranked by ELO or most recent match, success rate of specific teams, and even an AI agent that takes the parsed json whose data is placed in a DB to describe what works and what doesn’t when analyzing inputted matches. I originally made this in ChatGPT 5, but then moved to Claude desktop Sonnet 4.5 using the desktop directory plug-in (that was expensive and I was always running into context window issues that forced me to create a new chats. It wasn’t until I discovered cursor that I had that aha moment and was able to complete my dashboard without issue. Now, I want to continue to develop this dashboard, but I have this crazy idea of wanting to train two AI bots locally to play each other through the battle simulator environment. I have tested it out a bit, but this was with ChatGPT 5 web browser being the source of the code. I just finished watching the claude opus 4.5 video and I was wondering if this would be a good use case for it and what would be the most efficient/least headache way of tapping into it and with what IDE. Apologies for the noob question, I’m learning at 100 mph off of just YouTube videos and I figured some human feedback would give some insight into what I might be missing out on.
