AppleGenie

Inspiration

Our inspiration for AppleGenie stemmed from the need to streamline and enhance customer support for Apple users. Recognizing the growing importance of quick and accurate responses in customer service, we aimed to create a solution that leverages AI to assist Apple Customer Agents in providing top-notch support efficiently.

What it does

AppleGenie is an innovative AI-driven platform designed to assist Apple Customer Agents in responding to customer inquiries effectively. By using a high-quality dataset of frequent Apple customer questions and optimal responses, the platform delivers precise and timely answers. This not only boosts the productivity of Apple employees but also ensures customers receive the best possible support experience.

How we built it

We built AppleGenie using Python libraries such as Langchain, OpenAI ChatGPT, and Streamlit to host our code. Our dataset, sourced from Kaggle, includes technical questions about Apple and their best-performing responses. The front-end was developed using HTML/CSS, and we implemented authentication through Auth0. By training a large language model (LLM) with embeddings, we vectorized the data and utilized prompt engineering to create a scalable solution.

Challenges we ran into

We faced several challenges, including learning about LLMs and data vectorization. Additionally, working as a team required us to master pair-programming and maintain frequent communication. One major hurdle was finding a collaborative space, which led us to adopt Git for cross-collaboration and frequent updates.

Accomplishments that we're proud of

We are proud of successfully creating a functional prototype of AppleGenie, especially since it was our first experience working with APIs and LLMs. Our effective use of communication and version control tools like Git was crucial in merging our work seamlessly.

What we learned

Through this project, we learned about the intricacies of AI and machine learning, particularly in the context of customer service. We also gained valuable experience in team collaboration, communication, and using Git for version control.

What's next for AppleGenie

The future of AppleGenie involves expanding its capabilities to handle a broader range of queries, including more advanced topics. We aim to train the platform to filter out easily answerable questions, allowing human agents to focus on more complex issues. This scalability can potentially lead to the development of more sophisticated language models, continually updated with new information, ultimately boosting the productivity of Apple’s customer service operations.