Descriptions:
In this episode of the Cognitive Revolution, host Nathan Labenz invites Daniel Miessler — security researcher and founder of Unsupervised Learning — to audit the personal AI infrastructure Nathan has built over the past several months, taking direct inspiration from Miessler’s earlier appearance on the show. The result is one of the most detailed public walkthroughs of a working “second brain” AI stack available anywhere.
Nathan’s setup has two distinct layers. The first is a Claude Code instance on his main laptop with access to a 1-gigabyte personal knowledge database covering five years of emails, calls, podcasts, social media posts, and DMs, enriched with monthly and annual summarization layers. The second layer consists of two autonomous AI agents — one Claude Code-based, one OpenAI-based, named Aidi and Clay — running on an always-on Mac Mini accessed remotely via Tailscale, Apple Screen Share, and a custom agent messaging app built by Claude Code itself. These agents have their own Gmail addresses, GitHub accounts, and Mercury virtual credit cards with strict spending limits.
Miessler offers pointed commentary on agent hierarchy design, security posture, and why clear authority structures currently outperform emergent teamwork in multi-agent systems. The episode also covers the practical realities of delegating scheduling to an AI agent — Nathan’s autonomous agents successfully booked a full week of podcast guests — and what it actually takes in terms of context-building investment to make such a system reliable.
📺 Source: Cognitive Revolution “How AI Changes Everything” · Published May 30, 2026
🏷️ Format: Podcast
