# 985: The Four Types of Memory Every AI Agent Needs, with Richmond Alake

An episode of [Super Data Science: ML & AI Podcast with Jon Krohn](https://invisiblenetwork.ai/podcast/1163599059), 21 Apr 2026, 1 h 04 min.

Audio: https://www.podtrac.com/pts/redirect.mp3/chrt.fm/track/E581B9/arttrk.com/p/VI4CS/pscrb.fm/rss/p/traffic.megaphone.fm/SUPERDATASCIENCEPTYLTD8335157348.mp3?updated=1776767215

Oracle’s Director of AI Developer Experience Richmond Alake returns to the show to talk to Jon Krohn about agent memory; the network of systems, models, databases and LLMs that enable AI agents to learn and adapt over time. Listen to the episode to hear about Richmond’s “100 Days of Agent Memory” initiative, retrieval-augmented generation’s (RAG) limitations with AI agents, the layers of the AI agent stack, and what makes the Oracle AI database so useful to developers. Additional materials: www.superdatascience.com/985 Interested in sponsoring a SuperDataScience Podcast episode? Email natalie@superdatascience.com for sponsorship information. In this episode you will learn: • • (03:15) What agent memory is and why it’s important • • (28:28) RAG’s limitations for AI agents • • (35:19) What matters in the AI agent stack beyond memory • • (41:34) Why memory was undervalued in the AI agent stack

Shows are listed from Apple Podcasts' charts and directory, and Podcast Index where it is used. Every show and episode belongs to its publisher and plays from the publisher's own feed; this site hosts none of them.