Rental Agreement Agent

Using Docusign's Navigator API we are constructing a rental agreement (focused but not limited) chat agent leveraging ollama to run a local large-language-model.

Comment

Inspiration

The inspiration for the Rental Agreement Agent came from the need to simplify and streamline the process of understanding complex rental agreements. By leveraging DocuSign's Navigator API and integrating it with a local large language model, we aimed to create a tool that could assist users in navigating the often confusing terms and conditions of rental contracts.

What it does

The Rental Agreement Agent is a web application that:

  • Allows users to upload rental agreements through DocuSign's API
  • Fetches and displays a list of agreements from the user's DocuSign account
  • Enables users to select specific agreements for analysis
  • Utilizes a local large language model (via Ollama) to answer questions about the selected agreement
  • Provides a chat interface for users to interact with the AI and get clarifications on rental terms

How we built it

We constructed the Rental Agreement Agent using:

  • DocuSign's Navigator API for document management and retrieval
  • Streamlit for creating the web application interface
  • Ollama to run a local large language model (Mistral) for natural language processing
  • Cloudflare and Cloudflare Tunnel for secure hosting and access
  • A Raspberry Pi to host the web application locally

Challenges we ran into

While building this project, we encountered challenges such as:

  • Integrating multiple APIs and services (DocuSign, Ollama, Cloudflare)
  • Ensuring secure authentication and data handling
  • Optimizing the performance of the large language model on a Raspberry Pi
  • Creating a user-friendly interface that simplifies complex document analysis

Accomplishments that we're proud of

Key accomplishments include:

  • Successfully integrating DocuSign's Navigator API with a local AI model
  • Creating a seamless user experience for uploading and analyzing rental agreements
  • Implementing a responsive chat interface for interacting with the AI assistant
  • Hosting the application securely on a Raspberry Pi using Cloudflare Tunnel

What we learned

Through this project, the team likely gained valuable insights into:

  • Working with DocuSign's API ecosystem
  • Implementing and fine-tuning large language models for specific use cases
  • Developing web applications with Streamlit
  • Setting up secure hosting solutions for sensitive document processing

What's next for Rental Agreement Agent

Future developments for the Rental Agreement Agent could include:

  • Expanding the AI's capabilities to handle a wider range of legal documents
  • Implementing multi-language support for international rental agreements
  • Enhancing the user interface with more interactive features and visualizations
  • Integrating with other document management platforms beyond DocuSign
  • Developing mobile applications for on-the-go access to the rental agreement assistant

Built With

Share this project:

Updates