How AI Is Built  cover image

#014 Building Predictable Agents through Prompting, Compression, and Memory Strategies

How AI Is Built

00:00

Memory Storage and Efficiency in AI Agents

The chapter explores the storage of past conversations and operational data in memory, utilizing Lang chain and MongoDB for integration and efficient data handling. It discusses strategies for optimizing performance, reducing costs, and improving efficiency in AI models through techniques like prompting compression and leveraging semantic cache. The speaker shares their preferred stack for building AI agents, mentions different frameworks for agentic workloads, and emphasizes the importance of experimentation in the field to find suitable tools and methods.

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app