Fannie Mae Loan Automation App

Inspiration

The inspiration behind our project stems from a deep desire to transform the home buying experience into something efficient, transparent, and empowering. We recognized the complexities and uncertainties that often surround mortgage applications, which can be daunting for both buyers and lenders.

Ultimately, our inspiration came from a vision of a more accessible, fair, and efficient home buying process for everyone involved. With our automation app, we hope to revolutionize the industry and help individuals confidently pursue their dreams of homeownership.

What it does

Our goal was to create a solution that simplifies the home buying experience by leveraging Streamlit's capabilities to make it user-friendly. We wanted to harness the potential of advanced algorithms to provide instant, data-driven loan approval decisions, reducing stress and uncertainty for applicants.

But we wanted to do more than just offer a binary outcome. We aimed to provide guidance and support to users, helping them understand how they can enhance their chances of securing a loan.

Moreover, we understood the importance of communication and transparency in this process. This realization led us to integrate a real-time chat feature, ensuring that users could seek clarification, ask questions, and build trust with lenders throughout their home buying journey.

How we built it

We built the Fannie Mae Loan Automation App using Streamlit, a fast, interactive, and open-source app framework for Machine Learning and Data Science projects. The backbone of our app is a cosine similarity model that computes a similarity matrix to find users with financial profiles resembling the input data. By integrating OpenAI's GPT-3.5 Turbo, we've also enabled real-time, AI-driven financial advice within the app, providing users with a virtual loan officer for personalized guidance.

Challenges we ran into

During the development phase, ensuring real-time responsiveness while handling large datasets was a challenge. Moreover, integrating the GPT-3.5 Turbo in a way that it understands the financial data context to provide accurate recommendations was a complex task. Debugging and optimizing the code to ensure seamless user experience required a diligent approach.

Accomplishments that we're proud of

We are particularly proud of the dynamic and interactive nature of our app, allowing users to either upload a dataset or input individual financial data. The seamless integration of GPT-3.5 Turbo has elevated our app to a new level, enabling it to provide intelligent, personalized financial advice. Moreover, the successful computation of similarity matrices to generate actionable recommendations showcases the robustness of our underlying machine learning model.

What we learned

Throughout this journey, we honed our skills in Streamlit, delved deeper into machine learning models, and gained a nuanced understanding of integrating AI with financial analysis. This project also provided us with valuable insights into optimizing performance and ensuring user-centric design to enhance the overall user experience.

What's next for Fannie Mae Loan Automation App

Looking ahead, we plan to incorporate more advanced machine learning models and broaden the range of financial scenarios covered by our app. Additionally, we aim to enhance the AI assistant's capabilities to provide more nuanced advice and explore collaborations with financial institutions to possibly integrate our app into their existing customer service platforms. Our objective is to continue refining the app to make financial literacy and guidance more accessible to everyone.

Built With

• cosine
• openai
• python
• sklearn
• streamlit