VectorDAO

What it does

The user lands on the VectorDAO platform. The user decides to contribute a PDF/CSV any type of file to the DAO for embedding. The user signs up or logs in to their VectorDAO account. The user navigates to the "Contribute" section of the platform through proposals and chats. Upon prompt, the user selects the "Embedding" contribution type. The user uploads the desired data from their device or from the Ocean Marketplace. The user provides necessary details or context about the data. The user submits the contribution for community review. Community members review the PDF, ask for clarifications, suggest improvements, or vote on its inclusion. Community can do a Side by Side testing of the main Vector Index and of the branch with the proposed dataset, just like a Github pull request. The dataset is approved for inclusion if it receives sufficient positive votes. The dataset is processed and embedded into the vectorstore, enhancing the AI's knowledge base. The user receives a notification of successful inclusion and any rewards offered by the DAO.

How we built it

VectorDAO: Shaping the Future of AI Chatbots

Inspiration Our journey started at the University of California, Berkeley, where we were deeply engaged in AI research. We witnessed firsthand the incredible potential of AI and the power of collective intelligence. As we delved into chatbot technology, it became clear that the traditional, centralized development approach wasn't doing justice to the diverse, multi-faceted nature of conversational AI. We were inspired by the open-source revolution led by platforms like GitHub and knew there had to be a similar, collaborative approach for AI chatbots. Thus, VectorDAO was born.

What We Learned This venture has been a continuous learning process. Through our research, we discovered the crucial role of an extensive, well-managed vectorstore in enhancing a chatbot's effectiveness. We also recognized the importance of diverse perspectives, realizing that the quality of a chatbot could significantly improve by sourcing input from a wider community.

Moreover, we realized that blockchain technology was more than just cryptocurrencies; it could provide secure, transparent, and decentralized solutions for managing and exchanging data.

How We Built It Leveraging the strengths of Orbis for social interactions and project management, Ocean Protocol for secure data exchange and minting, and Gnosis Safe for DAO creation, we created VectorDAO - a community-managed platform for AI chatbots.

Our approach was to integrate these powerful technologies into a cohesive, user-friendly platform. We prioritized collaborative development, enabling users to create branches of the original index, contribute, and even actively test modifications.

Challenges We Faced The integration of Ocean was a significant technical challenge, particularly due to the compatibility issues with the Mac M1 chip. This situation slowed our initial progress and forced us to reallocate our focus on the other aspects of VectorDAO development. Navigating this hurdle gave us a profound understanding of the intricacies of platform integration and led us to devise efficient problem-solving strategies.

Creating a DAO to govern this whole process added another layer of complexity, particularly in ensuring that decisions are made democratically and securely. However, every challenge we faced only served to strengthen our conviction in the need for VectorDAO.

Looking Ahead As we move forward, we're excited to see the impact of VectorDAO on the world of AI chatbots. We believe in the power of collective intelligence and can't wait to see what our community will create.

Built With

• ocean
• orbis