Salesforce Just Hired Seven AI Employees. They Already Have Names.

A friend of mine runs a mid-size Salesforce consultancy here in Austin. Last week he texted me: "Did you see what Salesforce just dropped? They named their agents. Like, actual human names."

He wasn't exaggerating. Salesforce rolled out seven AI agents last week, and they all have names. Casey handles customer service. Paige runs IT and HR tasks. Carter manages commerce. Hunter does sales (launching November). Marshall handles supply chain. Piper builds experience management. Fin does financial operations.

These aren't chatbots with personalities stapled on. Each one ships with pre-built skills and data connections into Salesforce's Customer 360. They also get a new "long-horizon runtime" that lets them work autonomously for days or weeks. Not minutes. Weeks.

That last part is the real story.

Most AI agents today operate in short bursts. You prompt them, they do a thing, they stop. Salesforce's long-horizon runtime changes that. Casey can run a customer retention campaign over two weeks, adjusting her approach based on response data, without anyone re-prompting her. Marshall can monitor supply chain disruptions and reroute orders over the course of a month.

This is the difference between hiring a temp for a single task and hiring someone who stays at the desk.

What this means if you sell AI solutions

If you're an AI agency or consultant building custom agents for clients, pay attention. Salesforce just gave enterprise buyers a reason to skip the custom build entirely.

The pitch used to be: "Your off-the-shelf tools can't do X, so let us build you something custom." That pitch gets harder when Salesforce ships seven agents with names, job descriptions, and pre-wired data access. A COO at a mid-market company can now point at Casey and say, "Why would I pay you $50K to build something I can turn on Tuesday?"

That's the threat. The opportunity is just as clear.

Configuration is where the money goes

These agents still need setup. They need specific business rules, the right data sources, and tuning for each company's workflows. Salesforce made deployment faster, not automatic.

Think of it like Shopify. Shopify killed the "let me build you an e-commerce site from scratch" business. But it created a massive market for Shopify consultants, theme customizers, and app developers. The same thing is happening with enterprise AI agents.

The agencies that thrive will be the ones who learn to work with these platforms, not against them. Build on top of Agentforce. Specialize in the configuration and integration layer. That's where the revenue moves.

The naming thing matters more than you think

Salesforce didn't name these agents to be cute. They named them because enterprises need to assign accountability. When a customer service issue falls through the cracks, someone needs to ask "what did Casey do?" instead of "what happened in the AI system?"

Named agents create organizational clarity. They make AI legible to the C-suite. A CEO doesn't want to hear about "the LLM pipeline." They want to know that Casey handled 4,200 tickets last month with a 94% resolution rate.

This is how AI ends up on the org chart.

So what do you do with this?

If you're a business owner on Salesforce, start evaluating which of these seven agents fits your operation. Customer service (Casey) and sales (Hunter, November launch) are the obvious starting points. The setup cost is far lower than building from scratch.

If you're an AI consultant or agency, stop treating enterprise platforms as the enemy. Learn Agentforce. Build specialization around the configuration, training, and integration gaps that Salesforce won't fill for individual customers.

The companies that move fastest will have these agents running real workflows by Q1. The ones still debating whether to build custom are going to find themselves explaining why their solution costs 10x more than the one with a name tag.

— Mark Garza, Laimen AI