Please describe your proposed solution

Our solution consists of two key components:

1. A forum-style platform that facilitates discussions for each dRep community, linked to specific governance actions.
2. A chatbot designed to onboard and orient users to the platform.

This builds upon our F12 funded project, AI Assistant for Governance: Empowering Team-Based dReps. In our ongoing work, we are training a large language model (LLM) to perform tasks that streamline dRep workflows, enhancing their efficiency. The new platform will integrate this AI dRep Agent, allowing all users to benefit from its knowledge and insights.

Our goal is to create an AI-powered platform that manages both dRep and delegate onboarding while providing a dedicated forum for communication on governance actions. The platform will feature an AI-driven chatbot to guide users through the initial onboarding process, helping them navigate and understand the Cardano governance framework.

Once onboarded, users will enter their personalized dRep space, where they can engage in discussions related to specific governance actions. The AI Agent will actively participate in these conversations, offering insights and facilitating dialogue. All users will be able to access and interact with different dRep channels; however, there will be a verification process to limit certain actions. Verified users, who have delegated to a dRep, will receive a checkmark when posting in their own dRep area, ensuring credibility and highlighting the relevance of their comments within the discussions.

Please define the positive impact your project will have on the wider Cardano community

The AI Agent LLM proposed for this project will be thoroughly trained in Cardano’s governance, including CIP-1694, as well as the inner workings of the Cardano blockchain. This includes expertise in Cardano’s Ouroboros protocols, parameters and their impact on the blockchain, the logistics of staking and delegating ADA, using and setting up various Cardano wallets, and running stake pools along with their development operations. By fine-tuning this LLM based on Meta’s Llama 3.2, it is already adept at understanding and communicating in multiple languages, making it accessible to a global audience.

The AI Agent's ability to interact with users from around the world, helping them understand the intricacies of Cardano governance and its ecosystem, will significantly reduce the learning curve for onboarding new users. This lowers barriers to entry for individuals and communities who wish to engage with Cardano, ultimately leading to more users for dApps and services built on the blockchain. Additionally, the platform addresses a key takeaway from the global CIP 1694 workshops: the need for a robust communication channel around governance actions, which this solution directly fulfills.

What is your capability to deliver your project with high levels of trust and accountability? How do you intend to validate if your approach is feasible?

This product will feature a front facing chat agent that directly interfaces with our AI LLM Model(agent) and other users.

One of our team members has already created an example or such a user interface, that interfaces with the Ollama Model engine, which we plan to leverage as a template that we will redesign for our needs in this effort.

This chat already processes and renders any type of markup that is input by the LLM or User. https://github.com/bakon11/ollama_react_webUI

What are the key milestones you need to achieve in order to complete your project successfully?

Milestone 1: Choose LLM to use to fine tune for our need, gather data for the training/fine tuning and how to host,

• Emphasis on future collaboration and partnership opportunities for higher load hosting
• Other option for hosting such as decentralised AI computing option

Milestone 2: List of AI functionalities:

• Design how users and LLM interact, what type of use cases.
• UI/UX design and Implementation, hosting of the front end implementation

Milestone 3: Blockchain Infra deployment:

• Deploying the Cardano stack and tools,
• Storing and updating on chain proposals and proposal progress
• Feeding the gathered onchain Data into the LLM

Milestone 4: LLM Deployment on scalable infra:

• Deploying the LLM Model on decentralized or distributed computing infrastructure that can support hundreds of users
• Doing internal testing Between UI and deployed LLM Model

Milestone 5: Beta Testing:

• Onboarding outside community members to help with testing
• Tweaking parameters of all infrastructure and software based on testing input.

Final Milestone: Launch and Project Completion

Who is in the project team and what are their roles?

Photrek Team

Administration

Megan Hess, Project Lead

Development/Technical

Maciej Nowosielski, Senior Developer

Igor Oliviera, Junior Developer

Raymond Mata, Systems Administrator

Please provide a cost breakdown of the proposed work and resources

Total Budget ₳ 100,000.00

Development: ₳ 45,000

Marketing: ₳ 10,000

Computing: ₳ 20,000

Program Management: ₳ 15,000

Contingency: ₳ 10,000

Yes, the platform will need to collect data from Govtools and other sources where governance actions are listed.

How does the cost of the project represent value for money for the Cardano ecosystem?

This solution represents value for money to the Cardano ecosystem by enhancing the onboarding process for users into Cardano governance. Moreover, the solution introduces an effective communication channel that directly addresses a critical community need identified in the CIP 1694 workshops. Ultimately, more user engagement will lead to increased participation and transaction activities on Cardano and ultimately will lead to ongoing development and innovation within the Cardano ecosystem.