Schedule-Driven AI Employees — Running No-Code SDRs on the Twin Platform

Jon Law opens by showing what an AI agent finished while he was asleep at 8 a.m. — 20 e-commerce companies discovered, founder contact info collected, personalized cold emails written and sent, the whole batch logged in a CRM, and a Slack summary delivered to him. He didn't touch any of it. The platform is called The Twin.

From there Jon sets up the problem. The 'AI chatbot vs workflow builder' debate misses the point. Chatbots are weak on context and memory; tools like N8N or Zapier eventually force you to be the engineer. What's needed is the middle — an agent that assembles steps itself, runs on a schedule, and keeps running while you're logged off. The video walks through what that implementation looks like in practice.

In the hands-on segment, the SDR agent outputs research on 20 companies into Google Sheets and then sends 20 personalized cold emails as part of the same flow. Each email is a properly structured full cold-email setup, and the picture of an AI SDR running full-time becomes concrete.

The next step is the time trigger. Set 'repeat once daily at 9 a.m.' and the whole thing flips to a fully automated flow. Even when the laptop closes, the agent keeps running in the cloud. Pricing is credit-based, not subscription-based, so nothing runs means nothing bills. Separate from Claude Tag's bid to be the organization-wide harness, this is the endpoint of a different track — small AI employees carved out at the level of a single business unit, stood up no-code.

Jon lays out the strengths and weaknesses. The strengths: schedule-driven reliability, and a much lighter technical-setup load compared to open-class agents. The weakness is credit pricing — the most valuable tasks tend to be complex and ambitious, but token and compute costs hit the bill directly.

His recommended starting point is 'automating simple tasks you do a lot.' A use case that absorbs 10 hours of manual work for roughly $30/month in tokens is the highest-ROI place to begin. Pricing-wise, free starter credits get you in, and the Pro plan adds $20/month of credits that only burn when the agent actually runs. The realistic path is to start AI employees 'cheap, narrow, and always-on,' measure value, and expand from there.