jdalton

Why is it when they can't make software solve a problem, they then basically blame the problem.

To think a simple hook could have prevent it.

Could be worse

You're mistaken.

A bong! Hilarious!!!

Except instead of taking years to build unicorns, they're getting $32 BILLION valuations with no product.

The OpenAI Mafia makes the PayPal Mafia look like a lemonade stand.

I'm sorry but if you're not letting Claude Code build your future you might be a dang fool.

Why spends countless hours for pride?

I'll consider this!

They all admit they don't understand how their models actually work, yet they want us to trust them with everything.

Then I realized what's really happening is they're racing for the biggest corporate prize in history...a potential $50 trillion valuation.

Are they playing us?

I gave up everything to escape poverty through code. Now I'm using AI to reclaim what I lost... time with my family.

What you doing with AI to reclaim your time?

I see it as net positive. Likely in most cases the meetings (or at least some of them) may not even be needed.

Dyslexia can also a pita :/

True that.

I hope MCP servers make this possible, but even they will require oauth :/

I believe you may be mischaracterizing the article's position. It didn't advocate for deploying without testing or any of the other items you mention. Rather, it discussed how well and quick AI solved a problem. There's an important distinction between what was actually proposed and the extreme position you're describing.

The art is certainly in the planning, which I think many programmers might struggle with. Think how many say things like "I don't write docblocks", "my code explains it all." They'll be the ones that might struggle with it the most.

I do agree with you on letting the AI write its own prompt. It's the better way to handle context, that way we learn how it wants to be prompted.

I try :)

This happens far too often. So many solutions are just a MRR money grab.

The client wants their problem solved, they weren't looking to solve the solutions provider's MRR indefinitely.

I provide AI samples of my work, with context, which it does pretty well.

I have these patterns part of the command so it's no buried deep inside a context window. Claude is great at looking at the entire codebase and following the styles and approaches it comes across.

I've lived the high stakes life. We can be this crazy passionate about being the HITL but that will eventually change.

Agreed.

I see those type of meeting just being labeled context gathering in the future.

It's sure as heck would make life easier.

I wouldn't be surprised if Jira offered that in the next 6 months. Wait until AI gets deeper into PM work..scrum could die and I wouldn't shed a tear.

Oh gosh, don't me started on AWS!

I don't think it said it wrote code flawlessly.

A centralized API key registry would be cool, but risky as hell.

It could possibly become that.

It would seem you're far off.

I asked Claude to integrate Google Indexing API. 34 seconds later, it was done. I hadn't even gotten the API keys by then. Crazy time to be alive.