quantium-ai/research — reverse-engineered prompt

Build me a small public research project for exploring unusual quantitative trading ideas in Python using Jupyter notebooks.

I want it to feel like a learning lab, not a production trading bot. Each notebook should test one market analysis or prediction idea, explain the intuition in plain English, load example price data, make clear charts, and show basic results so someone can understand whether the idea looks interesting. Include a mix of classic chart based ideas and modern machine learning style experiments, such as sequence models or language model inspired approaches, but keep everything educational and easy to follow.

Please add a clean README that explains the purpose of the project, says it is not financial advice, and makes it clear that predictions are not guaranteed. Set up the Python project so the notebooks run smoothly, with sensible dependencies and simple instructions. Look up current docs online if you need to.