We had @jevering, CEO of @GuildAI on the show to explain how companies are deploying AI agents, the challenges they're facing, and where Guild fits in the stack. "We saw firewalls and the cybersecurity software business arise because you had intentionally bad actors trying to compromise your company's infrastructure. Well, you have a new problem when you put AI agents inside your company. They're unintentionally bad. You can't really predict what they'll do, they're very goal-end oriented, and they don't know the difference between right and wrong. " "If you want to be able to tell what an agent did two years ago, you need to be able to look at a history and see what it touched. So you need auditability and logging. You might want an agent to be able to access your CRM but not access your finance system, so you need to set up policies for that." "Cost controls is another thing. You see a lot of people talking about the spend, token maxing, and how much agents are costing. But really the question is: how much are the agents adding value to your business? Can you look at each one and figure out if that agent is doing what it was set up to do? And that is a type of observability that you also need, and a system like ours provides it.”