Designing AI agents presents a significant challenge due to the stateless nature of large language models and their limited context windows. This necessitates meticulous engineering to ensure continuity and reliability in sequential interactions with LLMs. Effective performance requires rapid systems to store and retrieve short-term conversations, summaries, and long-term data.
Redis, an open-source, in-memory data store, is extensively used for high-performance caching, analytics, and message brokering. Recent developments have expanded Redis’ functionality to include vector search and semantic caching, making it an increasingly popular component of the agentic application stack.
Andrew Brookins, a Principal Applied AI Engineer at Redis, joins Sean Falconer to discuss building AI agents, the significance of memory in agents, hybrid search versus vector-only search, world models, and more.
Full Disclosure: This episode is sponsored by Redis.
Sean has been an academic, startup founder, and Googler, with published works on topics ranging from AI to quantum computing. Currently, he is an AI Entrepreneur in Residence at Confluent, focusing on AI strategy and thought leadership. Connect with Sean on LinkedIn.
Click here to see the transcript of this episode.
Sponsorship inquiries: [email protected]