Memory patterns & recipes
Battle-tested recipes for agent memory. Each pattern solves a real problem with concrete code.
Looking for where memory interfaces are heading? Explore our 20 Memory Interface Research explorations.
Single Agent Recall
Give a single AI agent persistent memory by storing learnings after tasks and injecting relevant memories before new ones. The foundational deja pattern.
Multi-Agent Shared Memory
Connect multiple AI agents through a single deja instance using scope isolation. Shared scope for cross-agent knowledge, agent-specific scopes for specialization.
Human-in-the-Loop Teaching
Let humans explicitly teach agents by saying 'remember this.' The agent stores the teaching with appropriate confidence and scope using deja's learn() API.
Agent-to-Agent via Shared Memory
Enable asynchronous agent-to-agent communication through shared deja memory. Agent A learns, Agent B injects -- no direct messaging required.
Inner Loop Steering
Use inject() at the start of every agent loop iteration to continuously steer behavior with relevant memories. Pre-task injection guides decisions in real time.
Incident Response Automation
Capture postmortem learnings in deja and automatically inject them during similar future incidents. Turn every outage into institutional memory.
Memory Hygiene
Keep deja memory clean and useful over time. Patterns for confidence decay, pruning stale memories, consolidating duplicates, and scheduled cleanup.