Inspiration
Inspired by the finesse of real-world sales experts, we've crafted the EshopCopilot to mirror the dynamic and intuitive flow of in-person sales interactions. Our goal is to provide users with a seamless and engaging experience that replicates the rapport and effectiveness of a human sales agent.
What it does
EshopCopilot initiates the user journey by building trust, showcasing deep knowledge in electric vehicles, and paying meticulous attention to the user's preferences. It conducts an interactive dialogue, asking pertinent questions to develop a comprehensive user profile. Leveraging a robust electric vehicle database, the Copilot proposes a selection of vehicles that align with the customer’s specific needs. Each recommendation comes with a detailed explanation of the choice and the distinguishing features of the recommended models, ensuring that the suggestions are customized based on the unique user profile.
How we built it
The architecture of our backend is constructed with Python and FastAPI, interfaced with the Claude 3 Haiku APIs for streamlined interactions. The data is systematically organized and securely housed on Azure Tables, providing a strong and reliable foundation for our application.
Challenges we ran into
The main challenge we encountered was organizing and sorting through extensive datasets to feed our AI model the structured data it required to function optimally and offer precise vehicle suggestions.
Accomplishments that we're proud of
We've successfully developed a user interface that is not only aesthetically pleasing but also universally accessible on web, Android, and iOS platforms. Our machine learning model is sophisticated, continuously amassing and weighing data to refine its recommendations for electric vehicles.
What we learned
This project has been a tremendous learning curve. Our team has excelled in collaboration and dedication, with each member going above and beyond, embracing new challenges with enthusiasm. We learned the importance of flexibility and mutual support in a fast-paced development environment, which has been invaluable for our project's success.
What's next for EshopCopilot - Salesforce X Mercedes
Moving forward, we aim to elevate EshopCopilot by refining the UI for even more intuitive interactions, reducing response times for a smoother user experience, and enhancing our machine learning algorithms for superior vehicle matching accuracy. Our collaboration with Salesforce and Mercedes will continue to drive innovation and user satisfaction in the electric vehicle marketplace.
Built With
- azure
- claude
- dart
- fastapi
- flutter
- javascript
- python
- streamlit
Log in or sign up for Devpost to join the conversation.