YouTube Captions On by Default

Every YouTube embed we ship turns captions on in the player. Long-form and podcast pages are why. Do not invent a track.

How AI Agents Fixed Our Power Hour Onboarding Automation

The Task Summary AI agents fixed our Power Hour onboarding automation after a client purchase exposed a broken flow. On June 30, 2026, Ryan White of Social Revelation bought a one-on-one Claude AI Agent Setup session after his Power Hour with Dennis Yu, and hit a dead thank-you page. Dennis asked Ops for a root cause analysis and a repeatable …

How we repurposed Optimize Live honors without a second hub

Unique phrase: OPTIMIZE-LIVE-HONOR-2026-08-26. This is the factory write-up for Dennis’s Optimize Live honor list — Jack Allard, Isaac Hall, and Marko Sipila — not a second leave-behind. The canonical talk page stays dennisyu.com/optimize. What was already there The Aug 18 leave-behind was live before this run. It already named Marko Sipila next to Cam Hazzard and Dylan Haugen. It did …

How to Manage Your Google Local Services Ads

Most LSA accounts aren’t underperforming because of budget. They’re underperforming because nobody is working them. Leads pile up unrated, photos stay stock, and the one report metric that actually predicts lead volume goes unwatched. Here’s the full process I use, start to finish. This assumes your account is already live. If it isn’t, start with setting up Local Service Ads …

Install stack: Python 3.11, mlx-vlm.server, official 4-bit Qwen3.8-27B, bound only to 127.0.0.1 port 8080.

How to Install Local Qwen (MLX, No Ollama)

Local Qwen on Apple Silicon is an MLX server on loopback: Python 3.11, official 4-bit Qwen3.8-27B, mlx_vlm.server on port 8080. Cursor stays the conductor. This install is the worker. Do not use Ollama or paste 127.0.0.1 into Cursor. Routing map: Cursor and local Qwen.