Crafting AI Agents with Persistent Memory - The Ultimate Guide.

Format:

Video

Language:

English

Physical Description:

1 online resource (239 minutes)

Place of Publication:

[Place of publication not identified] : PACKT Publishing, 2025.

Language Note:

In English.

System Details:

video file

Summary:

<b>Master the art of building AI agents that remember and evolve. From no-code to advanced frameworks like LangMem and Supabase, this hands-on guide is your complete toolkit.</b><h4>Key Features</h4><ul><li>Full-stack agent development with and without memory frameworks</li><li>Real-world projects using LangMem, Mem0, and Supabase</li><li>Visual explanations of memory concepts and architectures</li></ul><h4>Book Description</h4>The journey begins by laying a solid foundation in AI agent architecture and the critical role memory plays in enabling agents to reason, recall, and adapt. Early modules demystify the core principles of memory in AI, explaining types of memory and their use in cognitive experiences. Through a mix of clear visuals and conceptual overviews, learners understand why memory is no longer optional in next-gen agent design. As you progress, you'll build agents that persist memory without relying on any external frameworks—perfect for grasping how memory mechanics work from the ground up. These hands-on labs include complete coding walkthroughs and visual animations to solidify understanding. You’ll test, refine, and deploy your agents, gaining full-stack insight into memory-aware systems. Then, dive deep into cutting-edge frameworks like LangMem, Mem0, and Supabase. Each tool is explored through realistic use cases—from customer support bots to health coaches—highlighting how to structure, update, and scale memory using these libraries. Whether integrating LangChain’s cognitive agents or constructing memory layers via Supabase, you’ll learn to build robust agents that feel intelligent and alive. The capstone project ties everything together, delivering a production-grade health coach agent with full memory retention.<h4>What you will learn</h4><ul><li>Build AI agents with persistent, updatable memory</li><li>Design and implement memory architectures from scratch</li><li>Integrate LangMem and LangChain into intelligent agent workflows</li><li>Apply Mem0 and Supabase for scalable, open-source memory storage</li><li>Analyze memory models and choose appropriate strategies</li><li>Deploy and test agents in practical, real-world scenarios</li></ul><h4>Who this book is for</h4>This course is ideal for intermediate to advanced developers, AI engineers, and software professionals interested in agentic workflows. A working knowledge of Python and basic understanding of AI/LLM concepts is required. Familiarity with REST APIs and JSON is beneficial but not mandatory.

Notes:

Title from resource description page (viewed July 20, 2026).