therealdrag0's comments

On: “Car Wash” test with 53 models

[therealdrag0]

Surveys have floors due to mistakes, effort, and trolling

Reminds me of https://slatestarcodex.com/2020/05/28/bush-did-north-dakota/

On: I built Timeframe, our family e-paper dashboard

[therealdrag0]

Ya it’s so stupid, I’ve been wanting a display for over a decade and prices haven’t budged.

On: Minions: Stripe’s one-shot, end-to-end coding agents

[therealdrag0]

I wouldn’t say never. I spent the first 10 years of my career loving crafting code. Then I moved up in seniority and naturally my focus and prioritization had to shift. Even before AI I didn’t code that much, focusing more on design, planning, reviewing, firefighting, and team leadership (still an IC).

One exciting thing about AI is when I have an idea of something to do and can visualize it, instead of writing a ticket that sits in the backlog, I can use AI to vibe it up, with just a couple hours of my attention I can spare. Sometimes it works sometimes it doesn’t. But it’s fun and satisfying to get more shit done, and be able to scratch the same builder and solver itches in my 10% time.

On: How I use Claude Code: Separation of planning and execution

[therealdrag0]

Audit logging is different than developer logging… companies will have entire teams dedicated to audit systems.

On: How I use Claude Code: Separation of planning and execution

[therealdrag0]

For some software, sure but not most.

And you don’t blame humans anyways lol. Everywhere I’ve worked has had “blameless” postmortems. You don’t remove human review unless you have reasonable alternatives like high test coverage and other automated reviews.

[girvo → therealdrag0]

We still have performance reviews and are fired. There’s a human that is responsible.

“It’s AI all the way down” is either nonsense on its face, or the industry is dead already.

On: How I use Claude Code: Separation of planning and execution

[therealdrag0]

In 5 min you are one shotting smaller changes to the larger code base right? Not the entire 80k likes which was the other comments point afaict.

[Bishonen88 → therealdrag0]

Yeah, then I guess I misunderstood the post. Its smaller features one by one ofc.

On: The path to ubiquitous AI (17k tokens/sec)

[therealdrag0]

Do you not have experience with agents solving problems? They already successfully do this. They try different things until they get a solution.

On: SkillsBench: Benchmarking how well agent skills work across diverse tasks

[therealdrag0]

Modern science encourages publishing non-surprising results.

And also I’ve seen my manager LARP as an engineer by asking a model to generate a best practices doc for a service repo without supplying any additional context. So this sort of paper helps discourage that behavior.

On: Omega-3 is inversely related to risk of early-onset dementia

[therealdrag0]

I’d love it if we could ban single-study health papers from HN. Meta analysis or bust.

On: The AI boom is causing shortages everywhere else

[therealdrag0]

Yes why not?