​Claude Code for PMs — Think in Agents, Not Prompts

​You've used Claude Code. Maybe you've built a few things. But you're still mostly prompting — one task at a time, manually, when you remember to.

​This workshop teaches you the mental model for designing agent workflows as a PM:

​- How to spot agent-worthy work: which repetitive PM tasks are actually worth automating vs. which ones aren't

​- How to design multi-agent flows: when to split into separate agents, when to compose them, how to chain handoffs

​- How to add memory (Mem0): so your agents carry context across sessions instead of starting from zero every time

​- How to make agents run without you: triggers, schedules, and headless execution so things happen automatically

​We'll build a real working example live: but the goal isn't the specific agent. It's the framework you'll use to design your own.

​Prerequisites:

​You should have Claude Code installed and have used it at least a few times. This isn't a setup workshop — we're going straight to building.

​About me:

​I'm Alon Binman, CEO of Deway (and a technical PM).

​My entire daily operating system runs on Claude Code — multi-agent workflows handling meeting follow-ups, customer prep, pipeline reviews, and commitment tracking.

​Not a demo setup. The real thing, every day. I'll show you how it works under the hood.