We kicked off Zero to Agent in 30 Minutes this week with guest John Berryman, an AI consultant and contractor for Arcturus Labs. John has spent the past several years building AI products and consulting on how teams put them into production. He set the stage by defining an agent as a large language model wrapped in two loops. An outer loop passes messages back and forth to the user. (This is the basis of all AI chatbots.) What turns an AI tool from an assistant into an agent is the inner loop, which lets the model choose and run tools. This dual-loop structure “is really not that complicated,” John noted, and it hasn’t changed since 2023. What has changed are the tools and instructions available to agents, which have improved enough that teams can now build real products around this simple pattern expressed in natural language. The payoff for programming in natural language, he pointed out, is that subject matter experts can now read the instructions driving the AI, examine faulty responses to understand where the reasoning broke down, and make updates directly instead of relaying feedback to a product manager and an engineer.
A high-level approach for building an AI agent
To show what this looks like in practice, John demoed a review pipeline for job candidates, then broke the process down into a repeatable method for building AI products, which he calls “outside in.” Here’s how it works.
- Build the traditional software first. Start with the interface, the data model, and every piece of the application that doesn’t require AI. Define exactly what information the AI component needs as input and what it should produce as output.
- Fake the AI with a stub. Before writing any AI code, connect the interface to a stand-in that returns a static response. This confirms the rest of the system works before you introduce a model.
- Swap in a simple agent. Replace the stub with a real but minimal agent, which John built with Pydantic’s agent and an AI reviewer. Give it structured output requirements with validation to keep the model on track. John used three fields: update type, internal notes, and correspondence.
- Give the agent a small set of tools. John recommends starting with four capabilities: read, write, edit, and shell access. Models have learned bash and command-line tools during training, so this small toolkit lets the agent extend its own capabilities when necessary, such as running curl commands for research.
- Distill the intelligence into a skill using natural language. Instead of coding a state machine, write the agent’s context, decision criteria, and step-by-step workflow in plain English. Another tip from John: Build checklists into your skills so the agent confirms to itself that every step has been completed or fails fast when they haven’t.
John closed by predicting that agents are headed toward fewer purpose-built applications and more agents that work across tools and interfaces on a person’s behalf. He’ll continue that conversation in his session “Escaping the Harness” at the AI Superstream on July 23. It’s free to attend. Register here.
Coming up next week
If you’re still writing posts one at a time, next week’s episode will rewire how you think about content operations. Craig Hewitt, founder of Castos, will build a complete social media agent live using Hermes, the system architecture behind tools like OpenClaw. Join us live to see how Hermes handles the handoffs that turn an article into a full day of X, LinkedIn, or Instagram posts without manual prompting at each step, or catch up after the fact on YouTube, Spotify, Apple, or wherever you get your podcasts.
Ready to run models on your own terms? AI Codecon returns with three expert-packed hours on building with open source AI. Save your spot now.
