Agents at Work

[Music] Hi everyone. I'm Matt Lewis from the partnerships team at Verscell. And for the last few months, I've been working closely with the team at Slack around one big question. What would it look like if agents actually came to work with us? if they were our digital co-workers. Not just a feature, but they were part of the conversation and

systems we use already we use every day. Last week at Dreamforce, Verscell showed an early demo of this, our V0 agent working alongside you in Slack. Today, I want to show you how Versel built this, but more importantly, why you should build one, too. Because agents aren't just another feature. They represent a shift in how we get work done. But first, let's talk about how we got here.

Most of what we do at work is conversation, planning, deciding, aligning, sometimes debating. It's all language, but our tools don't understand language. We've spent decades translating it into code. We learned to speak to the computer, not because we wanted to, but because it was the only way to make it work. That was the trade we made. We made work faster, but only

by compressing our thoughts into code. And while that was necessary, agents can help flip that back around. The way we work today is already conversational. We just built tools that stopped listening. And that's where agents come in. Systems that can understand what we mean, not just what we click. So, let's start with language. Humans

have about 170,000 words to choose from. Computers, at least in JavaScript, have seven. That's the gap. We have 170,000 English words, seven JavaScript primitives. And when we talk, we use nuance, tone, context, emotion. When computers talk, they use syntax or code. And that's the

difference between how we express and how we compute. And for a while, we celebrated that simplification. It really was good. It made work faster, but it also meant narrower, describing rich human thought and smaller and smaller pieces of logic. So, let me show you what that compression looks like. Shall I compare thee to a summer's day?

Here's one of the most famous lines in the English language. And when Shakespeare wrote this, he wasn't defining logic. He was expressing emotion. and meaning and intent. But let's see how we would have to write this for a computer to understand it. So there are some of those JavaScript primitives I talked about before. And that's just the first line. So imagine trying to explain beauty or creativity

with an if statement. That's what we've built, a world where we turned our natural language into code. And now alongside this code, we built interfaces to talk to the computer. And these boxes represent some of those interfaces. We did this not because we wanted to, but because we had to. Each new type of input allowed us to tell the computer a