Intro

Listen now: Spotify, Apple and YouTube

What if you had a personalized AI toolkit—not just a chatbot—that actually remembered your projects, your workflows, and even your family’s preferences?

In this episode, Marc and Ben sit down with Mike Bal, a product leader experimenting at the frontier of AI tooling. Mike shares how he built a local memory system for Claude using Model Context Protocols (MCPs), enabling persistent knowledge graphs that connect everything from his product designs to his family’s vacation plans. They walk through how it works—step-by-step—including a live demo of Fleur (essentially a mini app marketplace to make it easy for non technical people to add MCPs to Claude), how Mike structures entities and relationships, and why this setup beats traditional RAG approaches for real-world usage.

If you’ve ever wanted your AI to truly understand you and the work you do—or you're curious how a product leader uses AI to streamline everything from design reviews to family logistics—this episode is packed with real-world inspiration and actionable examples.

All episodes of the podcast are also available on Spotify, Apple and YouTube.

New to the pod? Subscribe below to get the next episode in your inbox 👇

In this episode, we covered the following topics:

* (02:08) Why traditional RAG setups break down—and how knowledge graphs offer a more reliable foundation

* (6:55) What MCPs actually are and why they’re more like a toolkit than an API reference

* (11:45) Installing Fleur: the dead-simple way Mike connected Claude to a local memory system using MCPs

* (15:30) Why traditional APIs fall short—and how MCPs give AI tools structured, usable context

* (28:52) Live demo: How Claude pulls memory from a local knowledge graph—personal details, podcast history, and more

* (35:28) A real-world example of how Mike planned a family trip using AI memory, including dietary needs and activity preferences

* (44:10) How Claude’s memory MCP compares to project instructions and system prompts

* (52:32) How Mike uses memory graphs for PRD reviews, user pain points, and feature validation

* And more!

Links:

* Mike Bal: https://www.linkedin.com/in/mikebal/

* Mike’s Newsletter: https://www.leadingproduct.link/

* Mike’s YouTube: https://www.youtube.com/@thatmikebal

* David’s Bridal Wedding Planner Demo: https://pearl.davidsbridal.com/

* David’s Bridal’s Revival with AI: https://venturebeat.com/ai/retail-resurrection-davids-bridal-bets-its-future-on-ai-after-double-bankruptcy/

* Fleur MCP App Store: https://www.fleurmcp.com/

* Willow Voice (Get the first month free by using this link!): https://willowvoice.com/?ref=BEREZ64

* Marc: https://www.linkedin.com/in/marcbaselga/

* Ben: https://www.linkedin.com/in/benerez/

If you’re a new subscriber, we encourage you to check out some of the recent episodes you might have missed:

And here are some early favorites:

To support the podcast, please check out the links below:

* Preparing for a PM interview with tech companies like Meta and Google? Ben's course for Product Sense and Analytical Thinking PM interviews has been selected by Lenny Rachitsky as a Top Course in Product on Maven 🔥. While the next cohort won’t be until October, his popular AI Copilot for interview practice is available here, where you can also see a live demo (ChatGPT support coming soon).

* Supra has teamed up with Maven to bring you something special – courses that our own community members have personally curated. And because you're part of the Supra family, you get $100 off any of these handpicked selections with code SUPRAxMAVEN.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit suprainsider.substack.com