jatins

I used v1 and ofcourse it scarred me but even v2+ versions felt like they encouraged over engineering compared to alternatives which I had a easier time wrapping my head around

I guess the conclusion is Angular was the worst choice pre AI and continues to be post AI

I was genuinely asking because I think all harnesses are largely replaceable. They all are building the same thing.

I have used cc, oc, pi, shelley and codex fwiw

say more, what do you think CC does better?

For context: eMudhra pays[0] its software devs around 500 USD per month as salary

[0] https://x.com/shobhitic/status/2112875765570359475

Honestly the fact that LLMs can get 60% at this that most humans will get 0% is commendable for LLMs

Are you checking on desktop? On phone Argon one is completely broken

Yes, sometimes I wonder if the right interface was actually something like Github Pilot’s inline code completions rather than agentic loop.

But if your coworker is shipping 5x more PRs than you your manager couldn’t care less about understanding code

For open source projects which are free of high growth expectations that come with VC funding this might be the right move.

AI coding comes with the tradeoff that you lose understanding of your own code. That is an acceptable tradeoff when getting it out to users matters more than understanding code. But free of capitalistic pressures, why would you accept AI slop?

Replit is not acquired by Anthropic

If anything he is very pro-AI and far from a naysayer. Labelling anyone who doesn't isn't Linkedin style "stoked" "excited" about AI is not helpful.

This is a technology not like prior technologies. Are we okay if the technology discourages a whole generation of Mathematicians? If the technology leads to 10x fewer mathematicians -- what impact does that have on the field? These are the questions Tao is asking. And I don't think he himself claims to have all the answers, he just doesn't wanna see the math _community_ die.

flagged for being low value add slop

Congrats on this ship, it's hard to get anything out there!

However, as many other comments have mentioned -- how do you plan to get users in a crowded space?

Probably the most underwhelming feature of Chatgpt. I dont know why its so slow and overall terrible as compared to Claude artifacts

> superior approach

superior how -- What does it do better over Go channels in your opinion?

> We’ve made major improvements to the way Opus 5.5 writes and communicates, one of the most common areas of feedback we heard about Opus 5. Its messages are much easier to understand at a glance, which testers said helped during long working sessions.

Thank you.

This hits home. Had the _exact_ scenario play out last year at my last role

it works! others can stop the test now

What’s the goal here with posting AI comments on every post? Does it help you get a job, do you sell this account?

omp is slopware through and through. And it shows, slow as hell to use

> It's a fine project to do but clearly they put zero value on being familiar with the project's codebase/stack and ecosystem,

As much as I dislike it, I think that's the future of _all_ non-critical software (think social media, crms, CI, food delivery etc). Leadership in many companies is explicitly asking employees to have multiple agents running through the day and that will lead to this.

Read this for example: https://www.uber.com/in/en/blog/efficient-software-factory/ . A very useful system, I am sure. But when you have AI at every layer from code to review to triaging, rest assured AI is the only know who knows your system. And you better hope it's not telling you that something is load bearing during an incident.

Given that cost of building is down, strategy of most companies is turning into "do everything" and see what sticks

100% of this comment is AI, against HN guidelines

https://www.pangram.com/history/8597362a-878d-4548-afb6-30fa...

They backtested with public models, wouldn't the model weights already have the data? I double checked with chatgpt and looks like agents also had web search tool available so they could just lookup the past.

https://chatgpt.com/share/6aa00d5e-6920-83ee-8e3e-9cbf23f7bd...

Anytime I see an investment framework I look for a “Performance” section or at least a backtest

> clearly have a contrarian business strategy to the other AI labs

What’s contrarian? Dont they also sell API and subscription like every other lab?

Completely agreed. I think the distillation from Fable to Opus is doing well on benchmarks but not translating into day to day work

Opus 5 is remarkably bad at instruction following over long chats. I have to repeat “Be succint”, “talk like a friend or colleague would”, “no rambling” or some variant of it every few messages

I talk to Claude like no one is going to read it. full of typos, often via voice where I am literally just thinking out loud.

I don’t think anyone will find it valuable and I don’t want anyone to read it

The thing is beyond like top 10 AWS services(ec2, s3, sqs, aurora and few more) this is how most of AWS is. It’s just stuff thrown at the wall to see if it sticks.

I wouldn’t be surprised if median AWS service does less than 100k in annual revenue