jatins

What an unnecessarily negative and cynical reaction to what is neutral at worst and positive at best

Progress takes time and there can be mistakes is basically the story of our civilization tho

Yeah, no one is doubting it. People are just asking to not be lied to about how much progress has been made

As open source as their timeline algorithm?

I do wish I had more friends

All the signs of slop, short one line paragraphs.

Tested a passage on Pangram and 100% AI generated btw https://www.pangram.com/history/f6b20d8e-4eb4-4bc2-b325-ad39...

Why are you telling it that? No one is arguing LLMs are perfect but at this point dismissing them as stochastic parrots seems a bit silly

insane, unproven, unpredictable autocomplete math equation

It won IMO, solved Erdos problems. At what point will you stop saying that?

rewrites feel like an area where LLMs are better suited than humans imo

It’s mostly grunt work and LLMs are well suited for translation tasks (iirc transformers arch was originally invented for translation)

Extremely thorough and well written.

I was hoping it’d end in a “so how much did this cost?” so that others team looking at similar migrations have an estimate on what they can expect

oh wow, a blast from the past. I remember a site called staticdrop or statichost like a decade ago, before vercel and everything that did the same thing

Good to see great ideas making a comeback

friendly employees

Where are the employees? There are so few employees (other than cashier) on the ground in a football field sized warehouse. Good luck finding someone if you have a question

If I see this code in a new company I just joined I am quitting on day 1

Webernetes is intended to be used to make interactive Kubernetes content;

You think these people don't already have generational wealth?

Demis is practically a co-CEO at this point. If he leaves, that'd be a huge of trust in Google.

I think they'll catch up in pure model capabilities but they do such a terrible job of making products from which to use their models that having the best model doesn't end up mattering.

Is Gemini good at writing code? I am sure it is. But where is their Codex? And no, antigravity isn't it.

Is Gemini good at making visualizations? I am sure it is. But where are artifact or visualise skill in gemini.google.com similar to what's available on claude.ai?

What is an average user going to do raw model capability if the product surface isn't expressive enough?

two senior engineers in my org (in a FAANG) who vibe-code the most have lost literally all of their skills.

Unfortunately FAANG incentives this behaviour with their token leaderboards and general push for velocity over anything else (other than goog maybe)

Comparison never made sense to me

1/ When dealing with High level language I am not seeing assembly or the language it compiles to. It's not a leaky abstraction

2/ It's deterministic

The day my markdown file is the thing I deploy on AWS your analogy will stand

The short term pay for the lunacy of working there is not a sensible trade-off for decent engineers

Besides top 2 AI labs, don't they outpay everyone?

Yes, but hiring someone with no experience running an AI lab to run your AI lab is a terrible decision for CEO to make in the first place.

Well in 2026 most likely this step was also done by an agent with --dangerously-skip-permissions

Flagged for being AI generated garbage

HN guidelines explicitly prohibit AI generated comments

Don't post generated comments or AI-edited comments. HN is for conversation between humans.

Not sure what's the policy on AI posts but can't imagine anyone learning anything from that

In Japan, your corporate affiliation often dictates your social standing

Where does it not? As in which countries are not like this?

pretty much

Read through the intro and didn't understand a thing. Either I am dumb or this is dumb or both.

So no one's using Gemini on HN?

Depends on the task. If it's something that occurs a lot in training data like React/tailwind code then I don't think you need SOTA. Most reasoning models since Sonnet 3.5, Deepseek 3.1 et al will do fine for those tasks.

Doesn't justify 10x the cost in that case imo

Yeah, what's going on in these cos is that a PM is tasked to find ways of integrating AI into product and well if someone's payroll depends on it they _will_ find ways to integrate AI into product. "Hey I couldn't find ways to integrate AI into product" is not an acceptable response.

And it's not just Uber. My weather app has an AI weather summary these days

https://archive.is/QcIhS

I feel like I really need to learn how to raise money

Well, cofounding Github helps