Gold Quant Agent: AI-Powered Trading System Analyzing 6M Data Points

So I've just built this a gold coin agent that analyzes over six million data points on gold from 2004 to 2026. So every morning it runs styles and a back test using various strategies with over 20 years of price data and looks for volumes, liquidity pools and patterns that the human just wouldn't be able to do to this level. It works by running Python scripts and using a machine learning model called XG Boost. It then takes the trade signals and runs further tests on this to see if they were profitable. And then it runs this through a continuous self-improving loop using a super-based database where it documents every single trade that suggests on whether it's one or not calculating the wind rates to return the draw downs. So it constantly improves its performance. Currently I'm testing a liquidity suite strategy which is one of the most profitable and the XG Boost model is telling me that it has an accuracy of around 44 to 51% where they want to for us through all ratio. This is extremely profitable. And then I'll put these trade suggestions to the over-telegramed companies to buy, sell or hold. A system like this would just be unvisable for a manual trade as to do as it's working six million data points as using machine learning out. Has it a self-corrected cycle? So you want the full GitHub repo so you can go and copy this for yourself. Comment, comment below and I'll send it across to you.