ABOUT THIS EPISODE
AI gives you a solid draft, never a sharp one, because it doesn't know what you know. In this episode I unpack how I applied Andrej Karpathy's context wiki idea to my work as an AI-first PM: a vault of local markdown files that Claude reads as a second brain and I read in Obsidian. I cover what it does day to day (morning briefings, PRDs and prototypes drafted from real context, a live status hub), how to set it up without writing any code, the scheduled skills that stop it going stale, and where it still breaks down. Read the written version at itsjanhofmann.substack.com/p/context-is-everything-high-quality
English
United States
TRANSCRIPT 🔗
Are you the producer of this podcast?
Add a podcast transcript
Need Audio-to-Text?
Transcribe with Listen411 in Just 60 Seconds
OTHER EPISODES IN THIS PODCAST
In this episode, I reflect on my 5-month sabbatical during which I explored AI and re-discovered my creativity.
Ex-hacker, ex-Meta engineer and InboxAPI founder Shaon Diwakar joins me to talk AI agents, prompt injection, local models, and why Apple might win the AI race.
How do you use AI to get more done in 2026, personally and professionally? In Episode 2 of the Jan in the Loop Podcast, I break down the three big players (ChatGPT, Gemini and Claude), what's really changed with memory, context, skills, tools and agents, and the everyday ways I put AI to work, from…
I asked Claude to build me a D&D initiative tracker. An afternoon later I had a deployed campaign hub with a database, a session journal and an automated quest log, and I never wrote any code. Here's how it came together, what I'd do differently, and what it changed about how I build.
Disclaimer: The podcast and artwork embedded on this page are from Jan Hofmann, which is the property of its owner and not affiliated with or endorsed by Listen Notes, Inc.
EDIT
Thank you for helping to keep the podcast database up to date.
