Every couple of weeks, someone in a dev forum mentions in passing that they're on a $40/month Cursor Pro and Claude Code subscription. It has the same vibe as "I use Vim btw" — except now you're shelling out cold, hard cash.
I understand, I've been there. I had both subs running for a month because I'd rationalized that I was just being practical. But let's call a spade a spade.
The Setup
Cursor Pro is ±$20/month. Claude Code on the Pro plan is $20/month and goes up to $100/month or $200/month on the Max plan. Out of the starting gate, you're investing $40/month minimum in AI dev tooling (Pro plans), with the Max plan totaling between $120 – $220/month. And you haven't even opened a terminal yet.
The reasoning given is consistent each time. Cursor is excellent for inline autocomplete — that rapid, tab-tab-tab rhythm while you are coding. Claude Code is good at multi-step tasks that require agency — the sort of task where you say "refactor this module" and leave it to it. Different strokes for different folks, eh?
The Identity Play
I think something else is happening. Subscribing to both has become a status signal disguised as a productivity hack.
It's saying: "I write so much code, a single AI tool can't keep up." It's saying: "I'm serious about my trade enough to put my money where my mouth is." It's the equivalent of owning both a road bike and a gravel bike when you go for a ride twice a month.
The discourse around "I use both" follows a predictable pattern. Someone posts their setup. Others ask how they split usage. The original poster explains an elaborate system of when they reach for which tool. Everyone nods. Nobody asks the uncomfortable question: are you actually more productive?
Most Devs Don't Need Both
I'm not saying the tools aren't good. They are. But the overlap is enormous.
→ Both can generate code from natural language prompts
→ Both understand your codebase context
→ Both can refactor, explain, and debug
→ The "unique strengths" are real but narrow
For most developers working on most codebases, one tool covers 90% of the use cases. The remaining 10% is where the identity kicks in. You're not paying $20 extra for productivity. You're paying $20 extra to feel like the kind of developer who needs both.
I ended up canceling my second sub after a month. My output didn't change. My commit frequency didn't change. The only thing that changed was I stopped alt-tabbing between two AI tools trying to figure out which one should handle a given task. That decision overhead? It was costing me time, not saving it. 🤦
When It Actually Makes Sense
I'll give credit where it's due. There are real cases where dual-wielding works.
If you're doing heavy greenfield architecture work and writing lots of boilerplate in the same week, the split might genuinely help. If you're evaluating tools for a team purchase decision, running both is smart. If you're building developer tooling yourself and need to understand the landscape, sure.
But those are edge cases. Not the default. And definitely not worth turning into a personality trait. 😄
The Bigger Pattern
This article isn't really about Cursor or Claude Code. It's about why developers latch onto tools as identity markers rather than judge them as utilities.
We did it with editors. We did it with terminal setups. We did it with mechanical keyboards. Now we're doing it with AI subscriptions. The tool becomes the signal, and the signal becomes the point.
Pick the one that fits your workflow. Use it well. Save yourself the $20. Or don't — I'm not your financial advisor.
But if you find yourself explaining your "dual AI workflow" to someone at a meetup and you see their eyes glaze over, just be aware: you're the new "I use Arch btw" guy. And deep down, you knew it all along.
So here's my question: are you running multiple AI coding tools right now, and has it genuinely changed your output — or just your setup?
Top comments (0)