Creator
13:20Eric Siu runs the same four marketing prompts — a website build, a thumbnail redesign, a growth-strategy memo, and a batch of social clips — on GPT Sol 5.6 and Claude Fable 5, and picks a winner based on which one actually finishes the job.
July 14th
14:48A 14-minute breakdown of how to turn recurring AI workflows into compounding business systems, with a live AEO/SEO loop as the proof.
June 23rd
13:29A 13-minute inside look at the Skills-Evals-Loops system one agency uses to train non-engineers to do the work of four to ten people.
June 22nd
10:39A 10-minute framework walkthrough showing how to build a unified AI memory layer that compounds — instead of buying more tools that never talk to each other.
June 19th
12:56A 13-minute live screen-share of six Codex agent workflows actively generating revenue at Single Grain — running autonomously for days at a time.
June 18th
27:39A 27-minute conference keynote where a marketing agency founder shows the live agent architecture replacing his headcount — six named agents, one shared brain, $500K in attributed value from $2,500 in tokens.
June 15th
10:57An 11-minute case for why manually prompting AI agents is already dead — and what building business loops looks like in practice.
June 12th
07:42A 7-minute breakdown of five concrete revenue plays unlocked by the new Claude Opus 5 and Sonnet 5 models.
June 11th
51:06Alex Lieberman demos the 8-step Claude skill directory he spent 50 hours building — and shows how it cut a 30-hour post down to four.
June 3rd
11:19Eric Siu turns Claude Code's basic /goal slash command into an operator-grade revenue stack with overnight, night-queue, batch, and approval-gated autonomy.
May 19th