Financial Planning Agent with LangGraph

๐Ÿ“˜ Description

This project implements a financial planning agent using LangChain, LangGraph, and GPT-4o-mini. The assistant engages in natural conversations, interprets user intent, and helps with core financial tasks such as budgeting, savings goals, and net worth analysis. The system combines retrieval, prompt templating, and structured tool usage to provide intelligent, step-by-step financial guidance.

Developed as part of the MSDS 442 course at Northwestern University, this module emphasizes agentic reasoning, structured interaction flows, and transparency in automated financial decision support.

โธป

๐Ÿ”ง Features

  • Tool-Based Planning: Uses specialized tools for income tracking, expense summaries, and financial health analysis.
  • Multi-Step Dialogue: Handles extended financial queries through LangGraphโ€™s node-based orchestration.
  • User Prompt Interpretation: Translates casual prompts into structured goals and constraints (e.g., โ€œI want to save for a house in 3 yearsโ€).
  • Memory Checkpoints: Saves user interaction history for continuous planning across sessions.
  • Custom Output Formatting: Returns answers in user-friendly formats (e.g., goals, tables, summaries).

โธป

๐Ÿ’ก Key Insight

With the right structure, language models can guide users through real-world financial planning tasks by combining conversational input, formal logic, and persistent memory. This project shows how AI can shift from passive assistants to proactive planners.

๐Ÿ”— View the source code on GitHub