This project aims to enhance communication stability for remote workers.
It monitors and analyzes network issues using AI-driven diagnostics, then provides clear, step-by-step guidance to users without technical expertise.
The AI acts as a domain expert, handling complex technical analysis while enabling non-technical users to implement effective solutions easily.
@cyberbryanzhang, @dev-tomnorris, @fourold
📧 Questions? Email: cyber.bryanzhang@gmail.com
This project demonstrates a multi-agent AI system that monitors and diagnoses network performance in real time using NVIDIA Nemotron reasoning models.
It consists of:
- Network Data API (FastAPI) → Simulates real-time network metrics.
- Monitoring Agent → Observes network data and identifies anomalies.
- Diagnostic Agent → Determines the most probable root cause using Nemotron reasoning.
All agents communicate via local HTTP endpoints and share the same NVIDIA NIM API key.
+---------------------+ +--------------------+ +---------------------+
| Network Data API | ---> | Monitoring Agent | ---> | Diagnostic Agent |
| (FastAPI, port 8000)| | (Nemotron reasoning)| | (Nemotron reasoning)|
+---------------------+ +--------------------+ +---------------------+
^ |
| v
+-------------------< Shared .env (API key) >-------------+
- Python 3.10+ (✅ 3.13 works fine)
pip install fastapi uvicorn openai python-dotenv requestsuvicorn network_data:app --reload --port 8000
python diagnostic-agent.pypython nemotron-monitoring.pyFastAPI, NVIDIA Nemotron, Multi-Agent System, Network Diagnostics, AI Reasoning, Python 3.10+
🧠 AI Reasoning Engine | 🧩 Multi-Agent Architecture | ☁️ Network Stability for Remote Teams
Licensed under the Apache 2.0 License