FrontierAI Connect

FrontierAI Connect is a cutting-edge, AI-driven web platform designed to assist Frontier Communications customers in diagnosing their networking issues and discovering personalized product recommendations. The project leverages Next.js, Tailwind CSS, and shadcn/ui for the frontend, and FastAPI, Python, and SambaNova’s AI model for the backend. This integration provides a seamless experience, guiding customers through problem-solving steps and suggesting the most relevant internet and network-related products to enhance their service.

Our team developed this project as part of the HackUTD24 hackathon challenge.

Demo Video

FrontierAI-Connect-Demo.mp4

Project Goals & Reasoning

Our team’s main goal was to create an intelligent support system for Frontier Communications customers. Modern internet users often face complex network issues—slow speeds, poor Wi-Fi coverage, or security concerns—and may not know how to address them. By leveraging SambaNova's AI, we aim to:

1. Simplify Troubleshooting: Customers can describe their issues in plain language. The system interprets their problems and provides actionable advice.
2. Personalized Recommendations: Based on the customer’s network data, usage patterns, and stated issues, the AI recommends products tailored to their unique situation. For example, if a customer’s Wi-Fi signal is weak, the system might suggest an extender or a more robust fiber plan.
3. Reduce Support Load: Automating first-level troubleshooting and recommendations empowers customers to resolve issues quickly and reduces the burden on human support teams.
4. Enhance Customer Satisfaction: By providing meaningful, context-aware solutions, FrontierAI Connect aims to improve the overall customer experience and trust in the brand.

Technology Stack

• Frontend:
  • Next.js for server-side rendering and a smooth user experience.
  • Tailwind CSS for quick and responsive UI development.
  • shadcn/ui for enhanced and accessible UI components.
• Backend:
  • FastAPI for creating a performant, production-ready REST API.
  • Poetry for dependency management and virtual environment handling.
  • Pandas for data manipulation and analysis.
  • SambaNova / OpenAI-Compatible API for integrating with advanced AI models.
• Data Source:
  • Mock network and product data from CSV files (e.g., current_customers.csv).

Setup and Installation

Prerequisites

• Node.js (v14 or later) and npm/yarn for the frontend.

• Python 3.8+ for the backend.

• Poetry installed for Python dependency management:

  pip install poetry

• A valid SambaNova API key.

Backend Setup (FastAPI + Poetry)

1. Navigate to the Backend Directory:

   cd backend

2. Install Dependencies:

   poetry install

3. Create a .env File: In backend/src, create a .env file:

   touch backend/src/.env

   Add your SambaNova API key:

   SAMBA_API_KEY=your_sambanova_api_k
   

4. Data Files: Ensure data/current_customers.csv and other necessary CSV files are placed in the backend/data directory.

Frontend Setup (Next.js)

1. Navigate to the Frontend Directory:

   cd frontend

2. Install Dependencies:

   npm install

   or if you prefer Yarn:

   yarn install

Running the Application

Running the Backend

1. Activate the Poetry Environment:

   cd backend
   poetry shell
   

2. Run the Backend:

   uvicorn src.main:app --reload
   

   The backend should now be running at http://127.0.0.1:8000.

3. Test the Backend: Open http://127.0.0.1:8000/docs in your browser to view the swagger documentation and test the /recommendations endpoint.

Running the Frontend

1. Start the Frontend:

   cd frontend
   npm run dev

   or with Yarn:

   yarn dev

2. Access the Application: Open http://localhost:3000 in your browser to see the frontend.

License

This project is licensed under the MIT License.