Hacker News: oofbey

New comment by oofbey in "Google buys crashed airline Spirit's data at auction"

oofbey — Tue, 18 Aug 2026 15:49:55 +0000

They will definitely be selecting the lowest bidder for this. Or perhaps a more expensive bidder if they can find one whose proprietary scrubbing technology is “a half dozen regexes our intern thought up”.

New comment by oofbey in "Google is making private AI practical with homomorphic encryption"

oofbey — Sat, 15 Aug 2026 05:07:21 +0000

Former Googler here. E2EE is easy. Nobody gets promoted at Google for solving easy problems. In fact if you set out to solve an easy problem, it looks bad at performance review time.

New comment by oofbey in "Google is making private AI practical with homomorphic encryption"

oofbey — Sat, 15 Aug 2026 05:04:19 +0000

All true. HE will never be used for anything real because it’s way to slow and inefficient, meaning you can only run the stupidest models on it. And there’s no commercial incentive to make it work because collecting data is too valuable.

But at the same time HE is classic Google promotion fodder. Really difficult technical problem! That’s what gets people promoted at Google. Nobody actually cares about it or will use it? That also is what gets people promoted at Google. Because Google only employs “the smartest people on earth” which means people who solve difficult problems. Difficult problems - not problems that need solving.

New comment by oofbey in "The original URL for this prediction will no longer be available in 11 years (2011)"

oofbey — Sun, 09 Aug 2026 06:06:57 +0000

Yeah that was a clever bit of foresight there.

New comment by oofbey in "Can Intel finally beat ARM on performance per Watt?"

oofbey — Sat, 08 Aug 2026 19:26:05 +0000

Really curious to see how ByteDance’s 10T model works out.

New comment by oofbey in "Can Intel finally beat ARM on performance per Watt?"

oofbey — Sat, 08 Aug 2026 19:10:59 +0000

I think the Top500 benchmark is biased here. It’s a supercomputing benchmark which means it’s going to heavily weight SIMD and float64 performance, which are things that rarely matter for everyday computing. SIMD has uses in the real world - image or video editing, data crunching, graphics. But consumers basically never use fp64 any more, and so many modern chips have next to zero performance on them. That could easily skew the numbers in intel’s favor.

New comment by oofbey in "Prime Agent: A self-improving RLM agent"

oofbey — Thu, 06 Aug 2026 00:41:26 +0000

The core idea of the RLM paper is to make a regular LLM act more like a coding agent - offload context to something external that needs to be explicitly queried instead of filling up valuable context. The "recursion" part of the paper really only wins because they use a top-tier model for the root agent, and cheaper models for the sub-agents.

Prime Agent took the RLM idea (which is really just an academic view on how coding agents have always worked) and then added this "continual harness" idea. This part isn't super well described in the blog post, but includes some message passing between the agents, and the ability to share code.

Overall I chalk it up as neat, but not revolutionary. Another version of what most of these systems are already doing.

New comment by oofbey in "The front end framework for correctness: built on Effect, architected like Elm"

oofbey — Fri, 24 Jul 2026 15:56:37 +0000

As agentic coding matures, this kind of project is the right direction. The mechanics of writing the code become less important. But having a language framework that naturally resists mistakes will become increasingly useful and important.

Similar to how Rust is obviously a better choice today than C++. Who cares if the language is “more difficult” to code in, if the agents are doing the coding. What we need is building blocks that make it harder to author bugs.

New comment by oofbey in "The Kimi K3 Moment"

oofbey — Mon, 20 Jul 2026 05:45:10 +0000

That’s for the consumer app / chatbot. For the api the terms are different: https://platform.kimi.ai/docs/agreement/modeluse

New comment by oofbey in "Qwen 3.8"

oofbey — Mon, 20 Jul 2026 05:35:59 +0000

That depends. There are two wry different processes that both get called distillation. One is where you have a fully trained large model and you are converting it to a smaller model. That kind if vastly cheaper than training a full model. Sure you could do that in weeks maybe even days for a big model. But it requires you to have the weights of the teacher model.

The other kind of distillation is where you record the outputs from a teacher model and use it to train a smaller model from scratch. That kind of distillation is not so cheap. It’s cheaper than training a model fully from scratch - starting with pretraining, then alignment, RLHF, the whole riggamarole. But here you are still starting from nothing and need to figure out how to get trillions of random numbers aligned in a way that makes them act intelligent. This is still gonna take a very long time if you’re talking about trillions of parameters.

New comment by oofbey in "Qwen 3.8"

oofbey — Sun, 19 Jul 2026 16:57:08 +0000

These things take months to train. No chance this is a reaction to what just happened.

New comment by oofbey in "The Kimi K3 Moment"

oofbey — Sat, 18 Jul 2026 20:24:33 +0000

Correct: can't opt out of training. This is well documented.

"Can't use for commercial purposes" - incorrect AFAICT. In what sense do you mean this? The open weight MIT version obviously allows for commercial use, but I don't think that's what you're referring to, because training data is irrelevant on the open weight version. Pretty sure the API allows commercial use too. Maybe the free version doesn't? But who cares?

New comment by oofbey in "SWE-1.7 Reach Near GPT 5.5 and Opus Intelligence"

oofbey — Wed, 08 Jul 2026 20:29:52 +0000

Agreed on the likely mechanism. I'm not sure "overfitting" is even the right description. These things are of course absurdly complicated, and evaluating their quality down to a single number involves a lot of judgement and trade-offs. I think it's more "you get what you measure" which is true in human organizations too. Define a KPI and people work hard to make it go up, even if it's not quite right or has bad side-effects.

New comment by oofbey in "Frog-derived gut bacterium eradicates tumors in mice"

oofbey — Wed, 01 Jul 2026 16:28:54 +0000

It’s not that. The tumors they create in mice are just really fragile compared to natural tumors. Natural tumors that actually cause problems have probably grown for years and learned to avoid immune responses. They’re not densely packed and easy to get into all the parts.

This kind of treatment triggers an immune response, which the model tumors have never had to fight before. So it’s just an easy target.

New comment by oofbey in "Single Dose of Frog-Derived Gut Bacterium Eradicates 100% of Tumors in Mice"

oofbey — Wed, 01 Jul 2026 12:41:36 +0000

The blog is highly suspect, but the study is real. That said it’s not a big deal.

Curing cancer in a mouse model is not at all uncommon in new therapies. Mouse models like this are vastly easier to treat than real world cancer for a bunch of reasons. Fully curing mice is the baseline for a treatment to even be considered for further evaluation. And even then very few therapies end up succeeding in humans - low single digit percent.

So yes, another possible treatment. But not at all a breakthrough.

New comment by oofbey in "Previewing GPT‑5.6 Sol: a next-generation model"

oofbey — Fri, 26 Jun 2026 22:32:20 +0000

Another year, and OpenAI comes up with yet another naming scheme for their models. First it was integers (GPT2, GPT3). Then they added friendly names (remember Ada, Babbage, Curie, Davinci?), but decided against it. Instead we got dot integers (GPT3.5), then then letter-number modifiers (o1), plus word modifiers like o1-pro, o3-mini, or -mini-high, or codex, codex-max, Pro, etc.

Now they've got friendly cosmic names. And this time they want us to believe that this time they're gonna stick to a naming convention? I'll believe it when they do 3 releases in a row without inventing a new naming scheme.

Venezuela hit by 7.5 magnitude earthquake

oofbey — Thu, 25 Jun 2026 00:51:52 +0000

Article URL: https://apnews.com/article/venezuela-earthquake-caracas-7179acaee70a9c543f953852f15d4814

Comments URL: https://news.ycombinator.com/item?id=48667402

Points: 10

# Comments: 0

New comment by oofbey in "Developers don't understand CORS (2019)"

oofbey — Sun, 21 Jun 2026 05:09:21 +0000

As somebody who has spent a lot more than 10 minutes trying to figure out why CORS was blocking what seemed legitimate, I sympathize with people doing the wrong thing, and disagree with your assertion that it’s not that complicated. Maybe I’m just slow. But objectively I know I’m not.

New comment by oofbey in "Ask HN: Has anyone replaced Claude/GPT with a local model for daily coding?"

oofbey — Tue, 16 Jun 2026 05:55:35 +0000

I think there’s a reasonable argument that a burst bubble will cause prices to drop. Prices are very high because they’re trying to justify these trillion dollar valuations on IP alone. If that fantasy goes away then prices will fall down to just silicon and electricity, which looks more like Chinese model prices. Hard to say how it will play out but the direction isn’t obvious to me.

New comment by oofbey in "Rio de Janeiro's "homegrown" LLM appears to be a merge of an existing model"

oofbey — Sun, 14 Jun 2026 20:06:23 +0000

Correct. We used to think that because NN optimization is non-convex there are all these local minima. Now we know that once you get past the very early parts of training from random init, the loss surface is fairly smooth, and not really convex, but close enough in a bunch of ways - linear combinations of trained models are pretty much always valid combinations. You can think of fine tunings as deltas on the original model which can be summed together successfully. I think this paper first showed that to me: https://arxiv.org/pdf/1802.10026 which was 8 years ago now.