Agent Memory: Building Memory-Aware Agents
by DeepLearning.AI
Give agents persistent, multi-session memory with production-grade memory architectures.
Overview
Agent Memory: Building Memory-Aware Agents, a DeepLearning.AI short course created in partnership with Oracle and taught by Richmond Alake and Nacho Martínez, teaches how to design memory systems that give agents persistence, continuity, and the ability to learn across multiple sessions. Rather than treating memory as a bolt-on, the course follows a memory-first architecture in which long-term memory is external to the model, persistent, and structured. Across about two hours and seven lessons, learners build a Memory Manager for different memory types, implement semantic tool retrieval, construct memory extraction and consolidation pipelines, and create autonomous self-updating memory loops. It covers practical retrieval strategies—vector search, graph traversal, and relational queries—implemented with LangChain, Tavily, and Oracle AI Database as the agent memory core. The course is free to access, with quizzes, projects, and a certificate available through DeepLearning.AI Pro.
At a Glance
- Topic
- Agentic
- Level
- Intermediate
- Format
- Course
- Cost
- Freemium
- Duration
- ~2 hours, self-paced (7 lessons, 4 code examples)
- Provider
- DeepLearning.AI
- Hands-on
- Yes — code/exercises
- Certificate
- Available
What You’ll Learn
- ✓Design a memory-first agent architecture with persistent, structured long-term memory
- ✓Build a Memory Manager that handles multiple memory types
- ✓Implement memory extraction and consolidation pipelines
- ✓Add semantic tool retrieval and self-updating memory loops
- ✓Combine vector search, graph traversal, and relational queries for agent recall
Highlights
- •Built in partnership with Oracle; uses Oracle AI Database as the memory core
- •Hands-on with LangChain and Tavily
- •Focused specifically on the memory gap that breaks multi-session agents
- •Free to audit, with Pro extras and certificate
Who It’s For
Best For
- ✓Developers building agents that must persist state across sessions
- ✓AI engineers designing long-term memory for production agents
- ✓Teams moving past stateless, single-conversation agent demos
Prerequisites
- •Basic Python
- •Familiarity with LLM agents and retrieval concepts
FAQ
What is Agent Memory: Building Memory-Aware Agents?
A short, hands-on course for developers building AI agents that need to remember across sessions instead of resetting every conversation. Built with Oracle and taught by Richmond Alake and Nacho Martínez, it treats long-term memory as first-class agent infrastructure.
Is Agent Memory: Building Memory-Aware Agents free?
Agent Memory: Building Memory-Aware Agents offers free content, with paid options for certificates or premium features.
What level is Agent Memory: Building Memory-Aware Agents for?
Agent Memory: Building Memory-Aware Agents is aimed at a intermediate audience. Recommended background: Basic Python, Familiarity with LLM agents and retrieval concepts.
How long does Agent Memory: Building Memory-Aware Agents take?
Expect roughly ~2 hours, self-paced (7 lessons, 4 code examples). Most learners work through it at their own pace.
What will I learn from Agent Memory: Building Memory-Aware Agents?
You'll learn: Design a memory-first agent architecture with persistent, structured long-term memory; Build a Memory Manager that handles multiple memory types; Implement memory extraction and consolidation pipelines; Add semantic tool retrieval and self-updating memory loops; Combine vector search, graph traversal, and relational queries for agent recall.