Pumpkin Escobar

I was still using my copilot account, figured I’d see how the new pricing worked. I blew through 60% of my max+ limit in 1 day on absurdly light usage, promptly cancelled rather than upgrade from the $39 / month plan to the $100 /month plan.

I do think there’s a productivity help from AI but vibe coding everything is miserable and gives awful results. Targeted AI usage makes sense and I’ll refine my local AI usage and tooling for that.

It works pretty well for greenfield, that’s where I’ve been using it. A pretty involved monorepo with a decent bit of code and things are still going reasonably well.

Also, leaning on documents as deliverable artifacts for people and for AI awareness / consumption helps me.

Like put together an architecture document for either the entire application or specific parts and subsystems (or multiple documents with different granularity) those can list future considerations.

I pretty often will include those documents when generating proposals / specs to make sure the AI is building in the right direction. It’s not perfect, there are some rough edges and these are the early days of this whole process.

Openspec has an “explore” command you can use to give some thoughts and have agent spit out a high level plan, some back and forth, then create the proposal once you’re happy. I’ve also generated markdown documentation for a subsystem or thought so I can check it in, then continue improving that later, then when ready to turn it into a proposal and implement I start by referencing that document.

It sounds like a big part of what you’re talking about is just pre-feeding future concerns into one proposal for work that you plan to do in the future, also nothing wrong with that. I’ve used “in the future we will need … so design in that direction without implementing in this first iteration” sort of wording.

And not sure if this applies but I’ve found big monolithic code bases become too large for AI to work well in… enforcing some boundaries by breaking code into technical foundation and vertical slice functional modules really helps decouple code and focus on cross-module boundaries and interfaces, similarly to how they help reduce cognitive load for developers, helps for AI too

I don’t claim to have perfect memory and recall of all biases of all news sources but Ground news shows them slightly left of center and that’s my memory too. I will spend some time on deep introspection to figure out if I’m a fascist. Thanks.

Man, these “left leaning” papers and sites sure are carrying a lot of water trying to get republicans re-elected sharing this bullshit. The pushback is a lie meant to convince voters that they can vote republican and still sleep at night like those republicans will stand up to Trump.

Don’t fall for this bullshit and shame on these farces of journalism for getting played into or being complicit in doing PR for republicans re-election campaigns

Yeah right, fucking performance art. They won’t do shit agains fascist daddy

This article making it sound like Republican senators and house members were going to do anything to rein in Donald Trump sounds completely unbelievable. Most of the time the Democratic leadership barely makes any meaningful noise.

I’ll believe it when I see it.

As an American, all I have to say is…. Yep

Start “open-ish”, gain traction, start abusing market position, start closing things off, become hostile to your customers…

But those aren’t the only two options. For people who only care deeply about one issue, if neither candidate is on your side, many will just stay home.

That was intentional. They were trying to word the announcement to not make it sound like you’re now getting 1/5-1/9 as much AI for the same price.

Related: Microsoft and OpenAI announced end of exclusivity agreement today

I love that their email made it sound like it was a wash, like “we’re just changing our billing model, you’ll get credits now, samesies!” but then this pricing chart buried 3-levels down from the announcement lays out just how much less you’re going to get for the same price.

I wonder about all the startups who were bragging about their $10k/month AI coding bills being the best money they ever spent. When this new pricing kicks in and pushes it to $40k/month right around the time all the vibe-coded shit blows up their codebase, I wonder if they’ll still be so happy with their choices.

I interviewed for a place a while back and started asking about quality and velocity and how they balance it with AI developer tools, he said something like “One of our developers closed 400 PRs last month” and I instantly knew it was definitely not the place for me.

When you’re deranged they let you do it.

Maybe he needs a 2nd gold-plated jumbo jet?

and said there’s “no way” Republicans can win under these circumstances

Democratic leadership: Hold my beer

Somehow…

Behind a traefik reverse proxy with lets encrypt for ssl even though the services aren’t exposed to the internet?

Arch everywhere. LTS kernel on servers and zen kernel on desktop and laptop. I love the idea of nixos but in practice it felt like more work than it was worth (to me).

I originally did Debian on servers but after using arch for long enough and never having stability problems, it was easier to move to the same distro.

I think they’re saying before developers started using AI to write the code, not that there are ai features in vim.