MCP is basically just an RPC API that uses HTTP and JSON, with some other features useful for AI agents today.

The first is using agents locally to develop.

The second is developing an agent. Not necessarily for coding, mind you. Not even for just text sometimes.

They are different cases, MCP is great for the latter.

Flak spreads the damage better and does more kinetic damage than trying to ram a plane with a missile and hoping the concussion from a the resulting explosion damages something.

This was a sneak attack outside of an established war zone, for an illegal war, so don’t try to conflate this as an attack on America’s enemies. The USA made them their enemies themselves.

That’s not to say Claude sucks, but I think Cursor is really underrated and not well known. I think the IDE focus hurts them with non professional developers, but try using it the same as with Claude and you’ll be surprised, I bet. You can hook it up to GitHub and never touch the IDE if you want to.

1) devs who would have pushed shut code anyways

2) getting stress from above to push code faster with AI

Good devs will learn to use these tools effectively.

If you give a regular person a race car, they will crash it about as fast as their vibecoded app crashes. Give the same race car to a pro age it’s a different story.

I still think this was the right decision by the programming mods there. Talking about tools is pretty boring, and you need to train to use something like an LLM assistant. No one who can’t program a language should be using an LLM to learn it unless they know about 2-3 other languages already, IMO.

Software engineers who haven’t tried these tools don’t understand what they are, and vibe coders who never understood software are taking the mindshare in public because it sounds revolutionary to some and apocalyptic to others. You have to stop listening to the claw bros and try using these as tools yourself in small ways to see what it’s really about, IMO.

I think people here are thinking I’m building Gmail from scratch when I’m talking about adding additional database APIs and models in the style of stuff I’ve already done twenty times in the same application. That’s easy for even the dumbest LLM and verification is not more than a code review, maybe an hour or two of labor. It’s never perfect, but as stated elsewhere I know how to program this stuff already so I can just fix it on the spot before I commit it.

I understand though, some of you either can’t use agents to code because of one reason or another, or refuse to; both are valid. I’m saying that for my job programming in a specific industry for a specific application in a specific language that AI agents actually help me out more than hinder me, and I still ship quality code in the style that matches the code base.