Building a Private, Local AI Partner: A Navigator Podcast
A conversation about the Navigator project
If you'd rather listen than read, this 22-minute podcast offers an engaging discussion about Navigator, the challenges of building reliable AI systems, and the thinking behind the project.
The conversation explores why AI needs more than intelligence to be useful, how Navigator approaches problems of memory, continuity, privacy, and control, and what I'm learning along the way.
Listen to the Podcast
Runtime: 21 minutes, 39 seconds
Download Podcast (MP3, 16.4 MB)
About Navigator
Navigator is an experimental personal AI partner I'm developing on my own hardware, using locally hosted language models and software I've built to manage memory, context, and technical capabilities.
The goal isn't simply to create another chatbot. I want a system that remembers what's important, understands the context of previous conversations, helps me think through decisions, and can perform technical work within boundaries I control.
Building that system has been an interesting engineering challenge. Getting an AI model to produce an impressive result is relatively easy. Building a reliable system around it is the real work.
This podcast is a companion to my article, Building Navigator: An On-Premise AI Partner, the first installment in a three-part series about the project.
The podcast discussion and featured illustration were generated using AI.