Hey, welcome to the free archived edition of The Product Compass newsletter.
Every week, I share actionable tips, resources, and insights for PMs.
Here’s what you might have recently missed:
I Copied the Multi-Agent Research System by Anthropic. No Coding!
AI Agent Architectures: The Ultimate Guide With n8n Examples
Consider subscribing and upgrading your account, if you haven’t already, for the full experience:
Since Jan 2025, I’ve been promoting Lovable as the best no-code solution for building real, full-stack apps. In that time, the company has grown about 10 times from $6M to $60M ARR, becoming Europe’s fastest-growing AI startup.
But I’ve also emphasized one thing:
While working with Lovable doesn’t require coding, it does require basic technical knowledge.
As explained in best practices, a Lovable user should:
Review components Lovable creates,
Look at the database structure,
Understand Supabase RLS (row-level security) policies,
Know how to integrate Lovable with GitHub, Supabase, or Netlify.
Turned out, it was a problem for some readers.
Recently, someone in our community recommended Base44. I was skeptical. But I decided to test it.
Base44 turned out to be the simplest all-in-one solution for those who want to build prototypes, small full-stack apps, or even minimal products.
So, in today’s issue:
Demo: Building a RAG Chatbot With Base44
Base44 vs. Lovable: Feature Comparison
Conclusion
This post was not sponsored. The only sponsors that appeared here on Substack in 2025:
AI PM Certification by Miqdad Jaffer (Product Lead at OpenAI). I recently joined Miqdaq as AI Labs Build Leader.
AI Evals for Engineers and PMs by Hamel Husain (ML Engineer, 25 years of experience) and Shreya Shankar.
Those courses are currently the #1 and #2 ranked on Maven.
1. Demo: Building a RAG Chatbot With Base44
In 18 minutes, I built a RAG chatbot and recorded the entire process.
A full recording with everything I did step-by-step: