TechBearing — Practitioner OS

The problem

Operational knowledge for a practitioner was scattered across tools and lost between sessions. It needed to be queryable by AI agents against live data.

Architecture

A Go/Next.js monorepo of ~14 services. An MCP server over SSE exposes live operational data to agents; an ingestion layer keeps state current; a persistent-state design survives restarts. Self-hosted.

What I shipped

• ▹MCP server (SSE transport) consumed by AI agents
• ▹AI coaching model over live operational data
• ▹Automated multi-source ingestion pipelines
• ▹Persistent-state architecture, operated solo in production