TechPrufer
Get started
Decagon
Decagon logoDecagon · Feature trail

User Memory for AI Agents

1 post · Mar 2026 – Apr 2026

What this trail covers

Decagon is introducing User Memory, a system designed to provide AI agents with persistent context across multi-turn conversations and sessions. This capability captures conversational history and personal user signals, storing them as structured metadata attached to a user's profile. It aims to enable deeper personalization, surface user-specific insights for proactive engagement, and provide enterprise-grade governance with data portability and expiration controls. This complements existing sy

All Decagon trails
  1. User memory: Persistent context that makes every interaction feel personal | DecagonThis post introduces Decagon's User Memory system, which builds a user-level abstraction to capture and carry conversational context and personal signals across interactions. It complements existing CRMs and data warehouses by providing a dedicated layer for AI agents to access persistent, opt-in, and controllable customer data. The system allows for personalization by remembering conversation history and extracting structured metadata like feature requests and preferences. It also surfaces user-specific insights through a unified user profile and offers governance features like data portability, redaction, and expiration policies.
  2. Introducing Proactive Agents: User Memory and Outbound Voice built for the AI Concierge | DecagonThis post introduces Proactive Agents, which combines User Memory and Outbound Voice capabilities. User Memory is enhanced to persist conversational history across sessions and channels, capturing key signals like feature requests and sentiment as structured metadata. Outbound Voice enables AI agents to initiate calls proactively, handling real-world conditions like voicemail detection and regulatory compliance. Agent Workbench is also introduced as a tool for debugging and iterating on agent performance using logs and reasoning.