Staging environment

Claude Code & Cowork: Build Your One-Person Company Crew

Hosted by Nathan Wang

325 students

In this video

What you'll learn

Build three agents that hand work to each other

Research, draft, review. You brief the outcome, the crew does the passing, and you judge the result.

Start free, then climb: chat, Cowork, Claude Code

We build live in Claude chat so every seat can follow. Cowork and Claude Code are rungs two and three.

Leave with the crew running, not a demo

You keep the prompts, the file layout, and the pattern for adding a fourth agent when you need one.

Why this topic matters

You do not need a co-founder or a headcount plan. You need a crew. In 45 minutes we build three agents that work together: one researches, one drafts, one reviews. We start in Claude chat, which is free and needs no install, so every seat can follow along live. Then I show where Cowork and Claude Code take the same crew next. You leave with the three agents running and ready to use, and the pattern for adding a fourth.

You'll learn from

Nathan Wang

Founder, Ipsion AI & Soul AI · ex-Meta AI/AR, ex-Apple · 50,000+ students

I build AI agents that hold a job, and I teach teams to do the same.

At Ipsion AI I built AgentOS and ClawBot, which lets you hire AI agents like employees and run a one-person company. Ipsion runs on it: growth, engineering, research and publishing are staffed by agents.

Before that, four years at Meta as a Staff TPM owning the embedded and user OS for the AI/AR glasses, and four years at Apple as a Senior TPM on iPhone, Watch and AirPods. Before Apple, semiconductor engineering and a US patent. PhD in engineering.

I have taught 50,000+ students: 17 AI courses on Udemy across 50+ countries, a 24-lesson agent course for LinkedIn Learning, and live cohorts here on Maven. Many write no code at all. What I teach is how to brief an agent crew, judge what it builds, and ship something real.

See all products from Nate

Go deeper with a course

Ship a Real AI Product with a Self-Improving Agent, No Engineers Needed
Nathan Wang
View syllabus