LangChain PM Nathan Drezner Demonstrates Agent Integration with External Services
LangChain Product Manager Nathan Drezner demonstrates in a video how to build a custom HTTP channel for a Managed Deep Agent to receive inbound requests from external services like HubSpot, Salesforce, or texting providers. He walks through building Reflex, an expense-reporting agent that extracts data from receipt photos and stores it in durable memory.
ImportanceLocalEvidenceE2 unreplicatedWrite-upQuick
LangChain published the video "HTTP Channels for Managed Deep Agents" on September 30, 2026, presented by Product Manager Nathan Drezner. The resource targets developers who need AI agents to interact with external, uncontrolled services.
In the demonstration, Drezner builds Reflex, an expense-reporting agent that receives receipt photos via iMessage through Photon, extracts the data, and stores it in durable memory backed by Context Hub. The video covers writing verify and parse functions, managing thread IDs per user, sending replies with a post hook, and checking results in LangSmith.
The content also discusses when to choose HTTP channels versus building a custom UI on the LangGraph SDK. For engineers evaluating the integration paths of Managed Deep Agents, this video provides a concrete operational framework and scenario examples.