What AI-Native Looks Like

ffagnerbrack about 6 hours ago 20 commentsRead Article on fagnerbrack.com

⚡ Community Insights

Discussion Sentiment

42% Positive

Analyzed from 716 words in the discussion.

Discussion (20 Comments)Read Original on HackerNews

randusername•about 5 hours ago

I'm starting to notice AI power users communicating in a shorthand tech-speak I find unintelligible. I genuinely cannot follow the individual points in this article or what the overall arc is.

Is this something that was already part of web dev or is it new?

sdcfgy•about 4 hours ago

I find wine helps. Lots of wine.

bakugo•about 5 hours ago

It's not the users doing the communication, it's their AIs.

And you, as a human, were not meant to read it - you're supposed to instead feed it back into your own AI agent and have it create a groundbreaking new AI startup based on it. Give it your Cloudflare credentials, sit back, sip a latte, and within an hour or two you'll have a beautiful, fully functional website, complete with fancy fonts, 30 different "No X. No Y. Just Z" taglines, and a $20 monthly subscription, ready to show off to investors. You'll be a highly successful entrepreneur by the end of the week.

Ampersander•about 4 hours ago

Doing drugs sounds both more rewarding and productive.

jononor•about 2 hours ago

Might even be cheaper!

nharziro•about 5 hours ago

I didn't find her article hard to understand. Maybe you have a reading comprehension issue?

NichoPaolucci•about 1 hour ago

Does this implementation really require this robust of infra / setup?

The entire stack feels overkill for a repo that has 2 followers and 29 stars.

I also am probably not the target demographic though, I just download things that I want to preserve and completely forget about them in a folder called 0_archive.

I just feel like in 2013 some fella would have had this project running on an old laptop in his garage hooked up to FTP and the uptime would be comparable.

I also love when AI adds tidbits like this:

> and staging has to pass before prod deploys.

Thanks for the clarity on that, Claude, you read my mind.

luc_•about 5 hours ago

Seems more like AI-in-the-loop more than AI-native.

I feel like this post didn't teach me anything I didn't already know or have an intuition about.

fagnerbrack•about 5 hours ago

My workflow these days is basically ultracode + PRs. I do control the features that go but barely need to check the code due to the amount of checks: lint, TS, Local AI visual Model for visual check for screen frame comparison (Mac M5 48GB mem), Pixel/Screenshot regression test for basic UI, code coverage/c8 to avoid trash code without tests, etc. etc.

Essentially the tests are just for the AI to remember what they did (as in the description + break if they change smth unintentionally).

ofjcihen•about 5 hours ago

I think that’s because across experience levels there really isn’t much to teach regarding “how to use AI”.

The only grind is still just do you know the subject matter, tools and frameworks, etc. because that’s the knowledge that lets you look at the output and say “this is fine” or “this is absolutely the worst implementation I’ve ever seen try again but do X”.

In my experience adding on multiple layers of controls etc aside from basic skills is essentially as useful as a ring of salt and a live offering.

hollowturtle•about 5 hours ago

Looking for a job? This sounds like personal marketing speech rather than a blog worth reading, no new insights no nothing

monegator•about 5 hours ago

no shit, it's some medium backed shit website.

Wish i could filter ALL of these, not just those with the medium website

nitwit005•about 1 hour ago

> 99.9% uptime (CloudWatch, Feb to Sep, insert your real number)

Did we forget to insert the real number?

dieselgate•about 1 hour ago

I read the article and it sounds kinda-sorta good at face value but the more I read and look into readplace.com the less things make sense and the more "ick" grows. Oh well just a puff piece... It's a big pet peeve of mine when someone uses a "CEO" title where it's probably very likely there's no Board or any other officers (or is a side project).

Author comes off as diligent, intelligent and dedicated... but might gain some credibility by having real humans read and proof the copy.

mikert89•about 5 hours ago

Ai native is only a term for the present moment. There is still skill in using ai. We are fast approaching where you ask the ai, and it does the task, there is no wrangling it

mijoharas•about 5 hours ago

Getting a 502 error. I had to come to the comments to check whether that was the joke.

dkobia•about 5 hours ago

13,000 tests. That tells you everything you need to know. It doesn't matter how experienced you say you are, if AI is doing all the work including writing and running tests because I know no human is actually reviewing these. I've recently found tests written by Fable that don't actually do anything in a very large codebase. My point is not to advocate for doing any of this manually, but to highlight one of the problems brought about by pumping out so much code.

I hate to sound cynical, but I don't know if bragging about being AI-native is something any of us should be doing. We're really just pulling a lever on a slot machine and crossing our fingers.

_ink_•about 1 hour ago

> I've recently found tests written by Fable that don't actually do anything in a very large codebase.

tbf, I've seen these kind of tests as well at $dayjob, written by humans before gpt was a thing. That doesn't invalid your point. But it's still funny that we expect the machine to be perfect, before it can do something for us.

fagnerbrack•about 3 hours ago

You need to ensure the whole system is optimised for LLM. Getting human written code and trying to run like this will be a failure since you'll be missing essential guidelines particularly on the domain-specific design fundamentals that are acquired in the beginning in a human feedback loop before the automation really kicks in

ashg10297•about 5 hours ago

So what does it look like? This is a load of booster talk tuned to reel in the devs who appreciate *****.