Build a Medical Chart Coding Pipeline with Daimon, Claude, and Neo4j

Short summary

Daimon is a Go sidecar that auto-generates LLM tools for vector stores and graph databases from YAML config, eliminating infrastructure boilerplate. This tutorial demonstrates building a complete medical chart ICD-10 coding pipeline using Daimon, Claude Sonnet, Neo4j, and Qdrant—combining semantic search candidate identification with graph-based verification. Includes Docker setup, Python client code, and system prompts for clinical coding workflows.

• •Daimon auto-generates LLM tools from declarative YAML, removing the need to hand-write schemas, dispatchers, and session management
• •Tutorial walks through a medical ICD-10 coding pipeline: parallel semantic search, graph-based verification, confidence scoring
• •Complete working example: config, Docker services, Python client (~60 lines), and installation methods for all platforms

Generated with AI, which can make mistakes.