Platform Engineer by day, AI Agent Architect by night! With over a decade of Kubernetes orchestration under my belt, and a few more decades in Infrastructure Development I'm now diving deep into the fascinating world of Agentic AI systems.
I fell in love with Docker orchestration and Kubernetes for its self-healing, declarative magic ✨ There's something beautiful about infrastructure that thinks for itself. Now I'm taking that philosophy to the next level by building AI agents that can actually operate that infrastructure!
Currently experimenting with cutting-edge Agent-to-Agent (A2A) communication and Model Context Protocol (MCP) to revolutionize how we approach SRE and operations. Because why should humans be the bottleneck when AI can help our infrastructure truly heal itself?
Building a Multi-Agent System Architecture that transforms traditional DevOps workflows:
User Input Layer → AI Investigates → Autonomous Fix → PR Created → Root Cause Analysis -> Slack Notification
The Vision: Infrastructure that doesn't just self-heal, but self-improves through intelligent agent collaboration.
Tech Stack: Kagent Framework • Agent Gateway • A2A • MCP • Kubernetes • ArgoCD • LLM Provider
Artificial Intelligence Reliability Engineering — a discipline focused on building autonomous, self-healing, and self-improving infrastructure systems using intelligent agents.
experience:
infrastructure: "20+ years"
kubernetes: "10+ years"
ai: "2+ years"
approach: "Declarative, GitOps-first, Agentic Systems"
philosophy: "Infrastructure as Code, Everything as Code"- Kagent Framework - Cloud-native agentic AI
- Agent-to-Agent (A2A) - Multi-agent orchestration
- Model Context Protocol (MCP) - Standardized tool communication
- Ollama - Local LLM deployment and management
- AWS EKS - Design, configure, manage at scale
- Azure AKS - Full-stack cloud-native solutions
- Multi-cloud - Because vendor lock-in is so last decade
#!/bin/bash
# My daily drivers
python --version # For serious automation
bash --version # For quick wins and glue code
kubectl version # Obviously- Agent-to-Agent Communication protocols
- Autonomous incident resolution workflows
- Multi-modal AI for infrastructure monitoring
- Agent Gateway AI Control Data plane
- GitOps + AI integration patterns
- Kubernetes never going to stop
- Advanced LLM fine-tuning for ops scenarios
- Prompt engineering for technical domains
- AI safety in production environments
- Observability and Governance for AI agent systems
- 🐳 Docker containers deployed: Lost count after 100K
- ☸️ Kubernetes clusters managed: From tiny dev clusters to massive production beasts
- 🤖 Favorite AI quote: "The best infrastructure is the one that manages itself"
- 🔧 Philosophy: "Stop Starting, Start Finishing"
- ☕ Powered by: Coffee, curiosity, and the occasional Kubernetes debugging sessions at 3 AM
Interested in the future of autonomous infrastructure? Let's chat about:
- 🤖 Agentic AI in DevOps/SRE
- ☸️ Kubernetes at scale
- 🔄 Agent-to-Agent architectures
- 🚀 The intersection of AI and infrastructure
Find me here: