🤖 Emmanuel's Digital Twin

Ask questions about Emmanuel Awotwe's professional background using RAG (Retrieval-Augmented Generation)

đź”— MCP Server Information

Server Name: digital-twin-mcp-server

Version: 1.0.0

MCP Endpoint: https://emma-digital-twin-mcp.vercel.app/api/mcp

Tool Name: query_digital_twin

⚙️ Claude Desktop Setup

Add this configuration to your Claude Desktop config file:

{
  "mcpServers": {
    "digital-twin": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://emma-digital-twin-mcp.vercel.app/api/mcp"
      ]
    }
  }
}

Alternative Configuration (if above doesn't work):

{
  "mcpServers": {
    "digital-twin": {
      "command": "node",
      "args": [
        "-e",
        "const { spawn } = require('child_process'); const proc = spawn('npx', ['-y', 'mcp-remote', 'https://emma-digital-twin-mcp.vercel.app/api/mcp'], { stdio: 'inherit' }); proc.on('exit', process.exit);"
      ]
    }
  }
}

Config file location: %APPDATA%\Claude\claude_desktop_config.json (Windows)

đź§Ş Quick Test

Test the digital twin functionality using server actions:

Use “Populate Vector Database” to index the updated profile data in Upstash Vector for semantic search.

đź’ˇ Example Questions

  • • “Tell me about your internship at AUSBIZ Consulting”
  • • “What's your Food RAG project that saves 1,200 hours per year?”
  • • “Describe your experience with Python and AI/ML technologies”
  • • “What are your salary expectations and career goals?”
  • • “Tell me about your Masters in Business Analytics at UTS”
  • • “What's your experience with Upstash Vector and Groq API?”
  • • “Describe your accounting background at Newmont Ghana Gold”

🛠️ Technology Stack

Framework:

Next.js 15.5.3

Vector DB:

Upstash Vector

AI Model:

Groq LLaMA-3.1-8B

MCP Handler:

mcp-handler

UI:

ShadCN/UI + Tailwind

Validation:

Zod