Summary
Cole Medin explains how he keeps up with the constant flood of AI news without burning hours every day. His answer is an “AI content engine” that lives inside his personal second brain, tracking hundreds of sources and delivering a short daily digest tailored to what he actually cares about.
The video walks through how the system works end to end. It pulls from Reddit, YouTube, RSS feeds, Hacker News, GitHub repositories, arXiv and product changelogs such as Claude Code, using tools including Apify to return agent-friendly markdown. Posts are deduplicated in a central database, then run through a cheap first-pass filter that handles the obvious keep-or-drop decisions, with only uncertain items sent to Claude Haiku. In one run of 864 posts, the filter step cost about three cents and the Haiku step about 42 cents, under fifty cents per day. Filtering criteria include whether an item is on-topic, new and concrete, or just a toy demo.
Medin also introduces a free skill in his GitHub skills repository that guides a coding agent through building the same system. It interviews you about your sources, fetch methods and filtering preferences, and builds a customized version using lessons from more than six months of daily use. Viewers will come away with a practical blueprint for personal AI news triage.
📺 Source: Cole Medin · Published October 15, 2026
🏷️ Format: Workflow Case Study
