A node-based workflow tool as an agent runtime
Using n8n as the "tool provider" for an LLM agent. Sub-workflows become tools the model can call. Changing agent capabilities becomes a workflow edit, not a deploy.
ReadWriting
Notes on the craft.
Architecture write-ups, patterns I trust, mistakes I've paid for. 5 posts and counting.
Two MCP server patterns: OAuth passthrough vs identity-aware search
Building Model Context Protocol servers for two very different jobs — fronting a SaaS with an OAuth broker, and exposing an internal knowledge base to authenticated users.
ReadDurable log plus event bus: a notification service for AI agents
One service becomes the audit trail and the dispatcher. How persistence-before-fanout, HMAC replay protection, and presigned-URL attachment sync fit together.
ReadBuilding an agent orchestrator without a framework
A hand-rolled state machine for LLM-driven chat: intent routing, tool invocation, SSE streaming, and out-of-band completion — and why we skipped LangGraph.
ReadDesigning an intent classification service for an LLM assistant
The front-of-funnel decision — direct answer, clarify, or call a tool — and why building it as a small, dedicated service pays off.
Read