Research Repository

A collection of research, experiments, and sample implementations in AI/ML agents and decentralized computing.

Research Focus Areas

AI/ML Agents

• Autonomous agent architectures
• Agent-based system design
• Multi-agent collaboration
• Agent learning and adaptation
• Natural language processing for agents

Decentralized Computing

• Distributed systems architecture
• Blockchain integration
• IPFS storage solutions
• Decentralized identity and authentication
• Smart contract implementations

Project Structure

.
├── agent/                     # AI agent research and documentation
│   ├── AI_Billing_Agent_Pipeline.md    # Implementation pipeline for AI billing agents
│   └── Future_of_AI_Billing_Agent.md   # Future research directions
├── sample_code/              # Example implementations
│   ├── webhook/             # Webhook integration samples
│   ├── nebula_block_storage/ # Nebula Block storage example
│   ├── mcp_nebula_block/     # MCP server for Nebula Block GPU resources
│   └── team_billing/         # FastAPI team billing example
├── venv/                    # Python virtual environment
├── .gitignore              # Git ignore file
├── LICENSE                  # Project license
└── README.md               # Project documentation

Code Architecture

Webhook Server (sample_code/webhook/)

• FastAPI Application: Provides verification endpoints
• Key Endpoints:
  • /webhook/verify-user: User quest verification
  • /webhook/developer/verify: Enhanced developer verification
  • /webhook/status/{verification_id}: Verification status check
  • /webhook/health: Health check
• Authentication: Bearer token verification
• Models:
  • VerificationRequest/Response: Basic verification
  • DeveloperVerificationRequest/Response: Extended verification

Team Billing API (sample_code/team_billing/)

• FastAPI Application: Team management and billing
• SQLModel ORM: Database operations
• Key Models:
  • Team: Team information
  • TeamMember: Team members
  • UsageRecord: Resource usage tracking
  • Invoice: Billing invoices
• Key Endpoints:
  • /teams/create: Team creation
  • /teams/invite: Member invitations
  • /billing/invoice/generate: Invoice generation
  • /usage/log: Usage tracking

Nebula Block Storage (sample_code/nebula_block_storage/)

• S3 Client: boto3-based storage operations
• Key Functions:
  • upload_file: File upload
  • download_file: File download
  • list_objects: Bucket contents listing
  • generate_presigned_url: Temporary access URLs
• Configuration: Environment-based setup

MCP Server (sample_code/mcp_nebula_block/)

• FastMCP Server: GPU resource management
• GPUInstance Model: GPU resource details
• NebulaGPUClient: Nebula Block API integration
• MCP Tools:
  • get_all_gpu_instances: All available GPUs
  • get_gpu_instances_by_region: Region-specific GPUs
  • get_gpu_instances_by_type: GPU type filtering

Research Samples

This repository contains various research samples and implementations, including:

1. AI Billing Agent Pipeline

   • Implementation of decentralized billing systems
   • Integration of AI agents with blockchain technology
   • Secure payment processing using cryptographic verification

2. Webhook Integration

   • Sample implementations for agent communication
   • Event-driven architecture examples
   • API integration patterns

3. Nebula Block Storage

   • Example implementation for Nebula Block object storage
   • S3-compatible API usage
   • File upload, download, and management operations
   • Based on Nebula Block Object Storage Documentation

4. Team Billing API

   • FastAPI application for managing teams, tracking usage, and generating invoices.
   • Uses SQLModel for database interactions (SQLite default).
   • Includes team creation, member invites, usage logging, and invoice generation/payment endpoints.
   • See sample_code/team_billing/ for details.

Getting Started

1. Clone the repository:

   git clone https://github.com/flyworker/research.git
   cd research

2. Set up the Python virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Unix/macOS
   # or
   .\venv\Scripts\activate  # On Windows

3. Install dependencies:

   pip install -r requirements.txt

Documentation

• AI Billing Agent Pipeline - Detailed implementation guide
• Future of AI Billing Agent - Research roadmap
• Nebula Block Storage Example - Storage implementation guide

Research Goals

This repository serves as a platform for:

• Exploring cutting-edge AI/ML agent architectures
• Developing decentralized computing solutions
• Sharing research findings and implementations
• Collaborating on innovative approaches to AI agent development

Contributing

Contributions are welcome! This is a research-focused repository, and we encourage:

• Sharing new research findings
• Contributing sample implementations
• Discussing and improving existing approaches
• Collaborating on new research directions

License

This project is licensed under the terms included in the LICENSE file.