Paul Barnhurst explains why AI speed matters less than correctness. He spawns multiple sub-agents to solve the same problem and check for consistency, catching impossible answers like negative 12,000. Running them in Python keeps costs low enough to verify at scale.
AI can build something quick, but if you want it to be right, often the majority of the time it is in check again. I don't think it changed anything from the checks. And there's a lot to be said for the fact that for both of us, and I got an advantage from this as well, like all the questions are multiple choice and the multiple choice are set so the answers are like, is it 100,004, 100,005, 100,006? Yeah, I've seen them before. If you get -12,000, you know you've done it wrong. And both of us can use that, right? So the AI was pretty sure. And I think Nico tried to spin off 1,000 subagents to try and check the work. And Claude said, No, that's stupid. You can have 20, and which is still overkill. That's a lot of sub-agents. It's an interesting point actually, because like before we were going on stage, like Nico was saying, I said, you know, I'm gonna have all of these sub-agents spinning off and checking my work. And I raised like a philosophical point of like, how fair is that? Because I don't get to have 4 other me's checking my work. But I suppose in reality there aren't 4 other me's, there is only one of me. So it's a fair comparison. Like you can copy the thing 10 times and all it does is cost you— it doesn't cost you 10 times as many credits because all of the sub-agents work exclusively in Python, so I think they're actually quite cheap.
Trust and Leadership Lessons for FP&A From 100+ Top Leaders with CFO Jon Laudie
