ripped_britches

7 ms·

61.
by ripped_britches 1y ago
It’s probably just that they started before gpt 5 was released. It’s a good judge.
62.
by ripped_britches 1y ago
Please for the love of god, if you work at cerebras, please put this on an API for me.
63.
by ripped_britches 1y ago
I’ve used sgpt and really liked that as prior art
64.
by ripped_britches 1y ago
Obviously by vibe reviewing it
65.
by ripped_britches 1y ago
Versus fleshy children, silicon children might be easier or harder to align because of profit interests. There could be a profit interest to make something very safe and beneficial. Or one to be extractive. In this case the shape of our mar
66.
by ripped_britches 1y ago
This is a horribly written article littered with logical fallacies. Nearly every point is just a random fact about gambling that doesn’t add to the argument. - fanduel owns networks. So what? - gambling taxes = corporate taxes in MD. This i
67.
by ripped_britches 1y ago
Amazing work! The UX is great. GPT-5-codex does more research before tackling a task, that is the biggest weakness for me not using Composer yet. Could you provide any color on whether ACP (from zed) will be supported?
68.
by ripped_britches 1y ago
What actually is this?
69.
by ripped_britches 1y ago
Apple and Google couldn’t take 15-30% that way
70.
by ripped_britches 1y ago
I make mistakes all the time. This seems like a genuine mistake, not malice. Imagine if you were talking about your own work online, you make an honest mistake, then the whole industry roasts you for it. I’m so tired of hearing everyone tak
71.
by ripped_britches 1y ago
That’s falsifiable quite easily by measuring tokens per second. Rather, the real reason codex takes longer is that it does more work to read more context. IMO the results are much better with codex, not even close
72.
by ripped_britches 1y ago
Yes mostly cursor for last 1.5 years but as of this month I am 100% codex CLI. Very freakin good
73.
by ripped_britches 1y ago
Sharing/public workflows seem like an obviously big security issue
74.
by ripped_britches 1y ago
Probably depends on what your business is and the LTV of your customers
75.
by ripped_britches 1y ago
Who could have guessed that servicing compliance requirements at scale would create such a good business model
76.
by ripped_britches 1y ago
All I can hope for is that the “effective context window” (some level before competency plummets) is like 1m+ tokens. I would give a finger to just put my entire codebase into a model every time I want to talk to it. For now I’m still only
77.
by ripped_britches 1y ago
“One persons map fixes everyone else’s” Hm somehow I feel like this is a giant step in the wrong direction.
78.
by ripped_britches 1y ago
Please comment under this thread if you have actually tried this and can compare it to another tool like Cursor, Codex, raw Claude, etc. I’m super not interested in hearing what people have to say from a distance without actually using it.
79.
by ripped_britches 1y ago
Then why are they paying for them? They can literally see the exact CPC and CPA / CPM ratios that justify their spending
80.
by ripped_britches 1y ago
I am a big fan of commercial advertising. I think it drives our global economy. However I think political and medical advertising need to be tightly regulated.
81.
by ripped_britches 1y ago
Please build patterns like manager/worker AI agent pairs. You spec a task and they work together on it in a loop, reviewing the code, etc.
82.
by ripped_britches 1y ago
We’re obviously heading towards a world where all training data is synthetic. What a compliance and legal risk otherwise.
83.
by ripped_britches 1y ago
Wow this is legitimately nuts
84.
by ripped_britches 1y ago
It’s just a joke setup for a funny story
85.
by ripped_britches 1y ago
It’s actually generally prime t-bones
86.
by ripped_britches 1y ago
Poor Jeff gonna miss that sweet AGI gravy
87.
by ripped_britches 1y ago
“Salesforce is just a database wrapper!” Boring take OpenAI has 700m WAU which is definitely not nothing
88.
by ripped_britches 1y ago
They’re nice if you let an agent decide what memories to store (so, schemaless in a way). This is how SOTA memory works today. There was a paper about it last month.
89.
by ripped_britches 1y ago
Picture the inverse of what you are saying. “Everyone can live 500 years but I think we should set up a program to randomly murder them at around 60-90 years old.”
90.
by ripped_britches 1y ago
HN’s response: “As someone neurotypical, I think this is completely wrong” Honestly I appreciated the one comment from someone with autism and everyone else complaining just seems like they think they know everything. One commenter even cal

More ›