Some business users spent ~30 minutes on an internal process, and we prototyped an "Agent" in Slack to take over. At first it didn't work, then it didn't work some more, eventually it ALMOST worked. Then one day, it worked, and the old business process died never to be revived.

Now it sits in a slack channel, and I watch it doing work, responding to ambiguity, and taking feedback/edits all day. It's unreal. It's literal magic. It saves a HUGE amount of time and gave us a pattern to do more.

This is the real deal. It's not easy to find problems with the right shape, and it's not easy to build agents that fit even when you do... but once it clicks, it clicks.

I had received four very different bids for a home repair project. Just wildly different itemization breakdowns, costs, timelines, scopes, even formats. Opus helped me turn it into an apples-apples comparison, filled in missing areas with reasonable inferences based on the other bids, provided a nice pdf I printed to review with my partner, even offered suggested key questions for follow up calls. It really clarified the advantages of one of the bids.

I use it professionally all the time and could cite technical scenarios where it’s become almost indispensable, but saving me time and money and reducing stress on this mundane stuff… now imagine applying to people’s stressors: job searches, health, big purchases, debt… there’s an opportunity to actually make people’s lives better. After 30 years of hype cycles, I should be wary of techno-optimism. But here I am feeling cautiously optimistic anyway.

I started to look at LLMs not as writing code, but rather as predicting what code it would expect someone to write given the context.

For some people that matches their expectation or they don't really have an expectation. While for other people it doesn't match their expectation.

When deepseek found a fix for a bug I couldn't find in minutes.

When deepseek again produced an entire web app that somewhat looked alright.

When Gemini could finally produce json was I specified.

The issue is, all LLMs can do. When they do, is boilerplate and code a mediocre coder could produce if they cared to try and insist.

In a way we should praise the ability of these things, but at what (in) efficiency. Code still need to be reviewed as we can't trust these things and context got a limit to entertain the idea of possibly having them fix their own mess.

MidJourney public discord channel.

The amount of masterpiece level art flowing per hour was astounding.

For every one doing a ninja waifu, there were ten doing art from davinci and leonardo crossed with hockney.

it almost gave you art sickness

Maybe my daily work is rather mundane compared to most people who frequent HN but I am able to create, think about, refine and then go through review cycles at least 2 or 3 times more quickly than I used to.

And software that I can imagine I might want to "make" or have at my fingertips is readily available even though I have a busy schedule with very little free time!

Also, I love feeling like a manager whose direct report actually does what I tell it to. Crazy good feeling.

When I tried, just for fun, to put together an MVP of a fully autonomous business, I wanted to see how far it would go, when I got it generally working to around a 30% level I stopped because it was enough to see people would make a concerted effort to build this for real. HN was not impressed, heh: https://news.ycombinator.com/item?id=44143928

Mine was testing out the copilot preview in the early days. Testing how well it knew semi obscure public codebases. Started filling out the first few lines and got the entire document word for word in tab complete.

That was the day I realised the plagiarism potential llms has.

I told the bot I liked Steely Dan, Eagles, Bob Seger, and Roxette and asked it for music recommendations. It replied with Toto. Exasperated, I wrote "Oh, shit, you stupid bot, you don't know ANYTHING about music!"

How could you not instantly see ChatGPT was absolutely revolutionary the first time you tried it? I was absolutely blown away and I'm sort of still are

That it could create mugshots of myself better than I could have managed to take.

Aka handsome, confident successful, affluent alpha male on a boat, yet looking perfectly like me.

January 2026 when i started using opus 4.5 and understood that it could do actual useful work beyond coding small snippets

A couple of years ago now.

I asked it to write a script that would search for a specific string in footers in a massive series of DOCX files and change them according to some rules. The strings ended up being embedded in cells within an invisible table in the footers, the LLM realized this and switched strategy to a full deep traversal of the underlying XML. It correctly processed like 50 of these files in about 10 minutes, using libraries I wasn't aware of. I had spent an hour being annoyed before trying.

It was an "oh shit" moment for at least that category of work.

Didn't had one yet. Apparently all I have is "crap, here we go again" whenever Claude is giving me a solution to the problem I am presenting to it. Because I understand where it goes and it's full of errors, but those are errors I can avoid. Together we cobble something in the end, I do learn something new as well, but was never "here is my prompt, then Claude delivered final solution next" - like so many commenters here point out they have.

Frankly, to an outsider whatever it presents looks legit, but as an expert I recognize its failures, which makes me even more entrenched in the idea to never use it outside my area of expertise.

I have a question for all them believers: If on a hypothetical scenario you, having no medical experience, find yourself and your child on a mountain, 12 hours away from nearest road, and your offspring is having appendicitis (let's assume your recognize this 100%), with a sharp knife and Claude at your disposal - would you risk to operate on your child? Or hurry the fuck down to get him to a hospital? I know I would chose to get him to a hospital, because that would be a better chance for my kid to live than me to operate on my kid with Claude's assistance. I am pretty sure I would kill my kid on that mountain. So yeah, outside my area of expertise I don't trust Claude one bit.

[flagged]

My first time using Grok. I'd been so used to using AI models that declined to do things I told them, like tagging people in a video feed, helping me "optimize" my taxes or managing my Twitter bot farm.

Grok just did these things for me, no questions asked, no ethical judgments. No woke.

Elon really doesn't get enough credit for Grok. People don't want the most powerful reasoning model or "constitutional AI". They just want a model that does what they say. Elon understood that insight (like he usually does) and no one else really did and that's probably why Grok has been growing rapidly over the last two years or so.