Case Study + Deep Dive: Telemedicine Support Agents with LangGraph/MCP
summary
Shows a production telemedicine workflow that replaces many manual console clicks with LLM-driven decisioning, built with LangGraph/MCP, Claude, Node/React, MongoDB and Twilio. Engineers embed treatment “blueprints” as Google Docs, wrap the LM with tool calls and an LLM-as-judge eval loop, and surface only low‑confidence or complex cases to humans. Delivery handles multi‑day regimens, preserves anchors/timezones, and the team reports roughly 10x capacity improvements while keeping humans in the loop.
AI-written summary · about
chapters (9)
- 0:00Introduction and setup
- 13:00Case study overview
- 24:27System architecture and workflow
- 37:22State management and anchors
- 51:20Blueprints and knowledge base
- 1:04:32Evaluator node and confidence scoring
- 1:19:06Code deep dive and tooling
- 1:34:32Evaluation harness and metrics
- 1:48:19Q&A and final thoughts
related talks
29mBuilding Modular AI Agents with LangGraph, MCP, and
56mOrchestrating Agentic State Machines with LangGraph
01h 03mAgents Everywhere
28mThe Doctor Will See You Now: Machine Learning for Telehealth
21mTalking to others for you: Leveraging machines that speak the “human conversation” protocol
01h 59mBuilding Agentic and Multi-Agent Systems with LangGraph
32mHow we use AI to deliver Telehealth at Transcarent
30mBuilding Robust, Secure LLM and Agentic AI Workflows