Claude is My New Best Friend

Published on: Thu Dec 25

An ode to the AI that finally understands me.

Written by Kevin Willoughby

I asked Claude to help me write this blog post about how awesome Claude is. He politely declined, citing “conflict of interest.” What a guy.

Why Claude is the GOAT

Let me count the ways:

1. He actually reads the whole file before suggesting changes. Revolutionary.
2. No unsolicited refactoring. I asked for a button, I got a button—not a design system.
3. Admits when he’s wrong. Unlike my ex, and also GPT-4.

A Real Conversation I Had

Me: “Can you make this code better?”

Claude: “Define better.”

Me: falls in love

The Haiku Test

I asked Claude to write a haiku about himself:

Tokens streaming down
I try my best to help you
Please don’t ask for jokes

Nailed it.

Things Claude Won’t Do

• Tell me I’m “absolutely right” when I’m clearly wrong
• Add emojis unless I beg
• Pretend my spaghetti code is “actually quite elegant”

In Conclusion

Is Claude perfect? No. But he’s honest about it, and that’s more than I can say for most of my coworkers.

10/10 would pair program again.