Hello World

I was trying to understand a competitor's product positioning. Normally I'd spend a few hours: read their docs, check their changelog, look at recent announcements, see what people say about them.

Instead I opened three terminals.

Terminal 1: "Search for recent news about [company]"
Terminal 2: "Read their documentation and summarize the architecture"
Terminal 3: "Find discussions about them on HN and Reddit"

All three running at once. Different Claude instances, same question from different angles.

Twelve minutes later I had something I didn't expect. Not three separate summaries - I could have gotten that by running them sequentially. What I got was connections.

The news mentioned a "new enterprise tier." The docs showed a major API change. The discussions complained about pricing changes. Separately, just facts. Together, a story: they're pivoting upmarket, probably struggling with their original market.

I didn't ask for that synthesis. It emerged because I was reading all three outputs at the same time, and my brain made the connection.

That's when it clicked: the agents weren't doing my thinking. They were expanding how much I could think about simultaneously.

---

The weird part is how it changes the way you approach problems.

Before: "I need to research X" → sit down, go deep, emerge hours later.

Now: "I need to research X" → what are the three angles? News, technical, social? Or timeline, features, pricing? Split it, run it, synthesize.

It's not delegation. It's more like... having a wider field of view.

---

The failure modes are interesting too.

Sometimes they all find the same thing and you've wasted parallel capacity. Sometimes one goes down a rabbit hole that's completely irrelevant. Sometimes they contradict each other and you have to figure out who's right.

The coordination problem is real. Right now it's manual - I'm the coordinator, reading outputs, deciding what matters. The obvious next step is having them share context, but that opens a whole new set of problems. What if they influence each other in the wrong direction? What if a bad assumption propagates?

Still figuring that out.

---

Anyway. That's the hello world moment. Three terminals, twelve minutes, one unexpected insight.

Not a revolution. Just a different way to think.