If you're building agents for clients, this Agent Operators conversation with Justin Bellware is worth working through. Justin explains how he combines client education with the Revenue Partner agents he builds. He starts by helping the owner understand what they want the system to do, then develops the workflow around that priority. A few things to look for in the episode: • Why he brought education into onboarding after selling larger packages • How he thinks about outbound, existing relationships, one-to-many opportunities, and content distribution • What goes into a campaign charter before the agent starts working • How writing, formatting, and editorial checks work together • Where his own experience shows up in the system Here's something to try after watching: Pick a task you know well. Write down the goal, the context someone would need, and the mistakes you would look for in their work. That is useful material to bring into an agent workflow, and a useful way to spot where the process still needs you. What task would you start with? Share it below, along with the one check you'd insist on before using the output. Full 22:34 episode attached. Justin's resources are in the first comment.

The new Agent Operators conversation with Corey Ganim and Spencer Kinney is ready. If you're deciding where AI fits in your business, these are the parts to start with: 01:10 — Feeling behind on AI. Corey and Spencer talk about the pressure to keep up. 08:57 — Managing agents. Corey shares his view on how work could shift from doing tasks to directing agents. 12:20 — Business value. Corey's three-part test: more money, time saved, or better product/service quality. 15:59 — Content and distribution. How Corey thinks about platform choice and testing ideas on X before expanding them. 20:42 — Skills that carry over. Selling, hiring, partnerships and positioning from e-commerce to AI consulting. Here's a useful way to watch: choose one recurring task in your business and apply Corey's three-part test to it. You don't need to have the automation built yet. Which outcome would you use to judge whether AI helped with that task, and what would you measure? Watch the full 25:57 episode attached to this post.

I recorded a 9:33 walkthrough of Bops, the business-ops agent app I built over a weekend. If you're building agents for yourself or clients, there are three parts worth looking at: 1. The actual task. I ask Boppy about a recent sales meeting and use email and company research for context. 2. The conversation. I move from the app to a phone call, then text, with the same context carried across. 3. The build. I show the Paper design, the T3 Code + Claude workflow, and how I connected computers, email, calls and business tools. As you watch, pick one task you'd hand to your own agent. List the information it needs, where the work happens, and how you'd check the result. Then compare that with the pieces in this build. What would that first task be for you? Full walkthrough in the first comment.

The new Agent Operators episode with Karsten Stinis is live. Karsten installs managed AI agents for businesses in Belgium. His offer is €2,750 for the setup, then €250 a month for support. The part I'd pay attention to is what happens during the installation: he watches the client work, finds a recurring task, and aims to have something useful waiting for them the next morning. Before you install the agent, decide what it will finish. A morning briefing is a good example: meetings, open tasks, and unanswered emails worth following up on. The person can open it, check the sources, and decide whether it helps. Whether you're building for yourself or a client, try scoping one overnight job: 1. The job: choose one recurring task. 2. The inputs: name the inbox, calendar, or task list it needs. Connect the accounts and check access. 3. The result: decide where the output should appear and when it should be ready. Make it easy to trace back to its sources. 4. The review: choose who checks it. Keep outgoing messages as drafts for the first run, and have a person approve them. Run it once. Read the output with the person who will use it. Use their corrections to improve the next run before adding another job. In the episode, Karsten walks through Hermes on Orgo, Composio for connecting business tools, Honcho for memory, and how he monitors usage and incidents. Keeping the workflow working is part of the service. He also reports one client's agent setup being built in 22 minutes. Humans still handle the tool and account connections. Watch the full conversation: How He Sells AI Agents for €2,750 + Monthly Support https://www.youtube.com/watch?v=Y8MJG3iDtSo What would you want your agent to have ready tomorrow morning? Share the job below.

Most multi-agent demos stop at "look, two agents can talk." In a client business, the hard part is knowing who owns the work, whether it was actually done, and what happens when an agent stalls. For Episode 3 of Agent Operators, I sat down with Jason Myers of AppLayer. He built Tango after realizing he had become the human copy-paste layer between his agents. We connected Hermes to a shared task list, watched workers claim jobs, and dug into audit trails, task leases, and recovery. Then I showed how Orgo cloud computers let you see and control the work when an agent needs a real desktop. In the conversation: • 04:19 — Hermes creates a task in Tango • 19:56 — Preventing two agents from doing the same job • 21:16 — Recovering stalled tasks • 30:25 — Agents working on Orgo cloud computers Watch the full 36-minute episode: https://youtu.be/LKCZXIGmrY4 If you’re deploying agents for clients, where do your handoffs still depend on you?