Most people use AI to chat. I built the infrastructure that lets AI act inside your own stack — trigger real automations, search real knowledge, write into real tools. Custom servers, self-hosted, everything under your own control.
MCP servers, n8n workflows, and a vault as the brain.
The full agentic infrastructure, built from the ground up. Custom MCP servers connect Claude Code to external APIs. n8n handles event-driven automation on my own VPS. An Obsidian-based Second Brain gives agents persistent memory. No wrappers, no SaaS.
An Obsidian vault that Claude can read, query, and write to.
The knowledge layer underneath everything. Markdown files with frontmatter that Claude searches semantically through a custom MCP tool. Decisions, learnings, skills — stored atomically and retrievable at any time. Input becomes usable context.
Custom MCP servers that turn any tool into a typed surface for Claude.
Multiple MCP servers — some built from scratch (Make.com, the Second Brain vault, a few Notion surfaces), others wired as clients (Asana, Linear, Figma, Stitch, Notebook LM, Excalidraw, Granola). Every one of them exposes its tools through Zod-validated schemas, so Claude can call them like typed functions instead of guessing at REST shapes.