How cfa-institute-rpc/the-automation-ahead works
Project type and purpose
• This is a monorepo of Python/Jupyter‐based code examples and tutorials illustrating generative AI applications in finance. It’s not a single full‐stack “app” but rather a set of loosely coupled “sub‐projects” (or modules) each demonstrating a workflow:
• Agents (agentic AI workflows and case studies)
• Introduction & Automation Framework (conceptual framework notebook)
• Practical Guide for LLMs (LLM comparison & execution notebooks)
• Retrieval Augmented Generation (RAG pipelines: proxy statements, news updater, visualizations)
• The goal: teach finance professionals how to wire up GenAI tools, data sources, and automation scripts.
File structure
• Top‐level
• .github/workflows: CI for daily briefs
• .env.example / .gitignore: environment and VCS config
• README.md, LICENSE
• Four main folders (Agents, Introduction and Automation Framework, Practical Guide For LLMs In the Financial Industry, Retrieval Augmented Generation)
• Modules
• Agents: Python packages + Jupyter case studies + “regime screening” and “sustainability research” workflows as DAG‐style code + a huge “reports” folder containing pre‐generated JSON/CSV outputs per ticker
• Introduction… & Practical Guide…: mostly standalone notebooks + markdown guides
• Retrieval Augmented Generation:
– Community Examples (static)
– Interfacing Proxy Statements: data folder of PDFs + evaluation notebook + ground truth
– Portfolio News Updater: automation_scripts folder with a daily_news_brief.py + RSS config + requirements.txt + env example
– Visualizations: notebooks demonstrating embedding plots + media files
README for additional insights
• Project is structured as a learning resource, not a hosted service
• Encourages forking/adapting scripts