MVP for a Retail Investment Advisor Powered by LLMs

1. Key Problems Tackled
2. Key Tools Used
3. Product Demo
4. Pitch Deck
5. MVP Screenshots

• Large amount of textual data
  • News data
  • Dense financial reports
• Large amount of numerical data
  • Financial time series are notoriously noisy
  • Requires good understanding of statistics & time series methods
• Quant finance models are gate kept
  • The average retail investor does not have the resources or know-how to utilize advanced recommendation & asset allocation models

• Python
• GPT (current LLM used in the product)
• langhcain (support with large context windows)
• streamlit (user interface)
• pandas/numpy/plotly/seaborn (data visualization)
• (TBD) scikit-learn/pytorch (needed in the future to suport dynamic quant models)