​

Business Requirements

In Agile project management, the concept of “Must have, Should have, Could have” is often used to prioritize requirements or features. Here is how it can be applied in a software development context:

1. Must have:
   • These are the essential requirements that are critical for the system to function properly or meet the core needs of the users. They are non-negotiable and must be included in the initial release.
2. Should have:
   • These requirements are important but not critical for the system to function. They enhance the user experience or provide additional value. They are prioritized after the Must-have items.
3. Could have:
   • These are nice-to-have features that are desirable but not necessary for the core functionality of the system. They can be considered for future releases or if time and resources permit.

By categorizing requirements into Must have, Should have, and Could have, the development team can focus on delivering the most critical features first while keeping the door open for additional enhancements based on priorities and constraints.

Here is how the requirements can be split into Must have, Should have, and Could have categories:

Must have:

• Meru AI must allow users to integrate the app to their specific Gmail account.
• Meru AI must automatically fetch and post email data with on-the-fly encryption.
• Meru AI must communicate with PaLM 2 or Gemini language model to process data and send it back to Gmail to reply or send new emails.
• Meru AI app must have an authenticated gated UI for its dashboard, analytics, and billing management for end-users.

Should have:

• Meru AI should support both Web2 and Web3 authentication and authorization-based access.
• Meru AI should provide a clear pricing and subscriptions-based pricing model with fiat and crypto options.

Could have:

• Meru AI could be able to translate, transcribe, and generate files based on emails that are being processed.
• Meru AI could provide the option for end-users to pick between PaLM 2 or Gemini or Bittensor subnets as the integrated model.
• Meru AI could provide users with staking and revenue share dashboard and analytics.
• Meru AI could provide users with a mobile application for both Android and iOS.
• Meru AI could store assets and files on a decentralized storage using IPFS & Filecoin.

By prioritizing the requirements in this manner, the development team can ensure that the essential features are implemented first, followed by important enhancements, and finally, considering additional features that could provide further value to the users.

​

Technical Requirements

Overview:

• Meru AI is a web application with decentralized layers accessible over the web.

Infrastructure:

• Meru AI should support infrastructure for multiple concurrent users hosted on Google Cloud using Google Compute Engine VMs with a load balancer (Cloud Load Balancing).

Email Processing:

• Email processing should run as background jobs that listen to new emails using push notifications from the Gmail API.

Environment Setup:

• Domains for Staging and Production environments should be managed by Cloudflare.
• Codebase maintainability will be on GitHub.

Tech Stack:

• Frontend: React, Next.js, React Native, Expo, TypeScript, Material UI.
• Backend: Node.js, Express.js, TypeScript, Redis, PostgreSQL, BullMQ.

Third-Party APIs:

• Third-party APIs to be used: Gmail API, PaLM API, Gemini API, Bittensor API, Stripe API.

Smart Contracts:

• Smart contracts will be written in Solidity, hosted on GitHub, and deployed on Goerli Ethereum Testnet and Ethereum Mainnet.

​

Main UML sequence diagram