Over the last 20 days, I started experimenting with a lightweight AI-assisted monitoring and operations framework called AIOps_Bot.
The goal was to explore how orchestration concepts, monitoring integrations, and AI-assisted workflows could be combined into a lightweight operational toolkit without requiring heavy infrastructure, installations, or external dependencies.
Current capabilities include:
• Unified alert visibility across multiple monitoring platforms
• SolarWinds SWQL automation workflows
• Multi-threaded IP monitoring operations
• SSH / Ping / Traceroute troubleshooting
• NCM config compare & change tracking
• Command runner workflows
• Health check and operational validation workflows
• Cisco DNAC, AppDynamics, and ThousandEyes integrations
• Mail/report generation workflows
• Structured JSON-driven orchestration model
The platform runs:
- without installation
- without additional dependency packages
- with standard user-level access
- as a lightweight portable executable
Architecturally, the project explores concepts similar to:
- AI-assisted observability
- orchestration-based operations
- RAG-style retrieval workflows
- MCP-style tool integration patterns
Still evolving and experimenting further, but this has been an interesting hands-on learning experience around observability, monitoring automation, orchestration, and AI-assisted operational workflows.
(Architecture diagram attached below.)
Note: Shared for architecture and technology discussion purposes only. No production or customer-sensitive information is included.
