I’m excited to announce the launch of a new limited-series podcast focused on AI workflows, automation, and how we’re all adapting to this rapidly changing landscape.

Why This Podcast? Link to heading

The pace of change has made the future foggy. My co-host Mya and I don’t know what it looks like, but we believe:

  • The proliferation of autonomous agents will make it even more important to have authoritative and verifiable sources of truth.
  • The mechanisms we use to share context with and collaborate with LLMs will be one of the most productive areas for advancement within the next 1-2 years.
  • When applied effectively, AI agents reduce tedium and increase the time that we get to spend on creative higher-level thinking.

We’re openly sharing our learning journey and how we’re adapting to these changes.

Episode 1: Now Available Link to heading

The first episode is now available on YouTube where we dive into the foundations of AI workflows and set the stage for what’s to come.

Upcoming Episodes Link to heading

Here’s our upcoming schedule (typically Fridays at 2:00 Central / 3:00 Eastern):

Episode 2 - Workflows & One-shots Link to heading

Can AI Workflows produce clean, actionable structured data? In this episode we’ll build on our progress from episode 1, building AI automation workflows based around a personal database. We’ll integrate FreshRSS, Huginn, n8n and baserow with MCPs and Claude Desktop. Then we’ll let coding agents lose to create one-shot UIs for interacting with the data.

Episode 3 - Agents & Assistants Link to heading

Is chat the only interface you need? Maybe. In this episode we’ll extend our workflows into agentic chat interfaces - and maybe even add voice chat!

Episode 4 - Homelab & Nix Link to heading

If it can be committed to git, it can be used as context for an AI. It turns out that LLMs + declarative configuration are a powerful combo. In this episode we’ll dive further into the “text-based-life” and learn how Josh manages his homelab with NixOS.

Episode 5 - Modern 12 Factor Link to heading

Heroku is back on the scene with an update to their 12-factor app for the cloud-native age. What does this really mean? How does it stand up in the face of LLM-based coding assistants?

Join the Journey Link to heading

Subscribe to follow along as we explore the intersection of AI, automation, and modern development practices. Each episode will include practical demonstrations, real-world applications, and honest discussions about what works and what doesn’t.

Stay tuned for more episodes starting this friday!

You can find all episodes on the videos page.