Introducing the Financial Freedom Agent.
Your tool to achieve financial freedom
Get insights on your spending habits, customized dashboards to help you stay on budget and see where you are overspending. Not everyone is a finance expert, when any question comes to mind you can ask your personalized financial advisor, all within the application.
A financial plan that works for you
Story
View your own financial trajectory
Debt always seems to build from nothing, here a debt growth chart forecasts what the benefits are of paying things off not, compared to pushing them off till next year.
Habits
View your spending habits
You know yourself better than anyone else, customize the categories of your transactions to categories that work for you
Budget
Get your customized breakdown overtime to see how you are tracking month by month
Customize your budget for whats right for you
Financial Advisor
Get personalized advice
Chat with the financial advisor who can help review your transactions and provide you personalized recommendations
Personalized Reminders
Using the trends in the data an email is sent on the day predicted to have the most out of budget spend.
The emails have both a text of tradiditonal advice, and a dynamically generated motivational text unique to each user.
Why this matters
Our mission
To enable financially struggling people to have the support and tools to be financially free.
Our Goal
Helping achieve financial freedom by leveraging the latest technology of generative AI agents
Financial independence is about having more choices.
60-70% of Americans are living paycheck to paycke
The inequality of wealth has substantially increased in the United States in recent decades. 60 - 70% of Americans are living paycheck to paycheck.
At the Edge: Unrecoverable Debt or Financial Freedom
A sudden change to economic status can quickly lead to a challenge and a majority of Americans are at the edge of going down a slippery slope of being in unrecoverable debt or achieving financial freedom. In a 2019 report on the economic well-being of U.S. households, the Federal Reserve Bank determined that nearly 40 percent of U.S. adults wouldn’t be able to cover a $400 emergency with cash, savings or a credit card charge that they could quickly pay off.
Money Choices and Emotional Decisions
Multiple studies have shown that behavior is the major reason that can help people living paycheck to paycheck. Per the book Psychology of Money, the premise is that doing well with money has a little to do with how smart you are and a lot to do with how you behave.
How We Built It
Models
Classification Model
Revolutionizing Financial Insights: Our advanced classification model leverages NLP and text processing to categorize credit card transactions into 8 distinct areas with unparalleled precision. Achieving over 90% accuracy in transaction prediction, our solution offers users a clear view of their spending habits and curates recommendations to enhance their financial well-being.
Persona Clustering Model
Empowering Personalized Financial Journeys: Our innovative k-means clustering model discerns unique user personas based on behaviors, revolutionizing the way we support low-income individuals. By tailoring experiences and providing personalized analytics and support, we guide users toward achieving their spending and budgeting aspirations.
Personalized Budget Tool
Automate Budgeting Experience: The budget tool leverages an analytical model based on the outcomes of persona clustering and classification model. For each user group (outcome of persona clustering model), we analyze historical transaction data to determine the average spending percentage across different categories. This outcome, combined with an individual's spending history, enables us to craft personalized budget recommendations for each user.
Financial Agent
Personalized AI Assistance: The financial agent is powered by AWS Bedrock, ensuring top-tier data security. It utilizes the RAG technique and an enhanced prompt system for precise, personalized responses to users' financial queries. The Claude Instance model is preferred for its cost-efficiency and rapid response times.
Application Architecture
Application
Streamlit
AWS Services
Bedrock Agent
Cloud 9
RDS PostgreSQL
S3
Lambda
EventBridge
Simple Email Service
The Team
Ashrita Brahmaroutu
Classification Modeling and Data Analysis Lead
Steven Cherry
Front end Web App Lead
Kodzai Nyakurimwa
Infrastructure and Data Engineering Co-Lead
Andrew Sandico
Product and Project Lead
Steven Tong
NLP, LLM, and Agent Lead
Angelina Wedemeyer
Infrastructure and Data Engineering Co-Lead
Special Thanks to:
Thank you to our instructors and peers at UC Berkeley MIDS.
A special callout to professor Jennifer Zhu and the capstone instructors Joyce Chen and Kira Wetzel.