Practical writing on AI automation engineering — how to build LangChain agents, design n8n workflows, implement RAG pipelines, and ship full-stack AI systems that work in production.
Written by Fasil Hawultie, AI Automation Engineer. Each article comes from real client work and production systems — not tutorials, but decisions made under real constraints.
How to build an intelligent model router for AI agents that classifies query complexity and picks the cheapest model that can handle it. Reduces token spend and latency without sacrificing quality. Includes a full n8n no-code walkthrough.
A clear breakdown of Model Context Protocol (MCP), traditional APIs, and Google's Agent-to-Agent (A2A) protocol. When to use each, how they differ, and what actually matters for building AI systems.
A practical guide to building a production RAG chatbot. Choosing the right models, comparing Pinecone vs Gemini FileSearch, optimizing latency, and deploying on real websites.