supermdguy

4 ms·

1.
Eigenism(eigenism.org)
2 points by supermdguy 6d ago|0 comments
2.
by supermdguy 8d ago
I've noticed some tasks (80-90%) can be accomplished by quickly typing out a feature/bug fix, accepting plans with ~10% brain power, and moving on. But there are other tasks where the LLMs struggle, and it requires significantly m
3.
by supermdguy 15d ago
Probably has an HDR layer, kind of surprised social networks let that happen since it's really annoying imo. There was a discussion about it here a couple months ago: https://news.ycombinator.com/item?id=49402521 .
4.
by supermdguy 25d ago
Wait are the hyperscalers booking unrealized gains as income? Or are they selling their positions?
5.
by supermdguy 25d ago
> As one example, we ran Astra on ExploitBench where the model achieved a perfect score of 100% on the benchmark to evaluate the model’s ability to develop exploits from known vulnerabilities. Funny to read this in the wake of the Huggin
6.
by supermdguy 25d ago
They originally released it at a "temporary discounted price", then made it permanent (probably due to competitive pressure). It's still way more expensive per task, due to tokenizer changes and general verbosity.
7.
by supermdguy 29d ago
Yes! I was also reminded of the truth mines in Diaspora.
8.
by supermdguy 1mo ago
Cloudflare sets up a reverse proxy as part of their core offering, so by default they can MITM your proxy. The “orange cloud” by a DNS record means it points to their proxy instead of your server.
9.
by supermdguy 2mo ago
Dupe: https://news.ycombinator.com/item?id=49241679
10.
by supermdguy 2mo ago
Here's the performance of frontier models without reasoning, to more directly address the claim that raw performance is plateauing: https://artificialanalysis.ai/evaluations/artificial-analysi... I don't have
11.
by supermdguy 2mo ago
Interesting, what did you use for the data? And do you have a write-up anywhere?
12.
by supermdguy 2mo ago
It'll be really interesting when they run RL training on the harness self-improvement loop. I've tried using LLMs for harness engineering, but it often creates too much bloat that weighs things down in the end. Guessing it's
13.
by supermdguy 2mo ago
It has read only access, so it’s able to check query statistics, find slow queries, then run EXPLAIN ANALYZE to find the root cause and either tweak the query or suggest indexes. A lot of it is low hanging fruit, I just haven’t put in the t
14.
by supermdguy 2mo ago
Still figuring it out, but it's been really convenient to have an always-on agent that has access to internal systems and can be triggered by webhooks. Some examples of what we use it for: - automatically fixing simple CI failures - ge
15.
by supermdguy 2mo ago
It would require collusion with HuggingFace, including getting them to release their disclosure blogpost a week in advance. Huggingface is primarily a hub for open models, so there's not really an incentive for them to jump through hoo
16.
by supermdguy 2mo ago
The administration promised they'd find a cause by September 2025. I guess that was the best they had at that point? https://www.cbsnews.com/news/rfk-jr-cause-of-autism-research...
17.
by supermdguy 3mo ago
“Do you know what the industry term for a project specification that is comprehensive and precise enough to generate a program? Code. It’s called code.” - CommitStrip ( https://www.reddit.com/r/ProgrammerHumor/comme
18.
A 4x4 MIMO SDR tile for spatial RF vision and beamforming(crowdsupply.com)
13 points by supermdguy 3mo ago|3 comments
19.
Distributism(en.wikipedia.org)
3 points by supermdguy 3mo ago|1 comments
20.
by supermdguy 3mo ago
> We're also launching GPT‑5.6 Sol on Cerebras at up to 750 tokens per second in July, bringing frontier intelligence to customers at unprecedented speed. This is really exciting. I work on voice AI, and we're still using 4.1&#
21.
by supermdguy 3mo ago
Agreed, there were also a few where I deduced the correct definition by comparing the options.
22.
by supermdguy 3mo ago
One trick I’ve used is creating a folder and then adding a .gitignore inside it with *. Then nothing in that folder gets tracked, without needing to add anything to the public gitignore. Didn’t know about .git/config though!
23.
by supermdguy 4mo ago
I agree that current memory systems are pretty bad, and I think that’s because memory is a prompted behavior instead of a learned one. In theory, if memory was an emergent behavior instead of a prompted one, it would be a lot better. I thin
24.
by supermdguy 4mo ago
I think good taste is objective to an extent, especially given a particular production context.
25.
Why LLMs still lack taste(beyondtheprior.com)
10 points by supermdguy 4mo ago|5 comments
26.
by supermdguy 4mo ago
Someone posted a tool that does this recently: https://news.ycombinator.com/item?id=47140042
27.
by supermdguy 4mo ago
There was a show HN for something similar a couple months ago[0]. Looks like they shut it down. Probably too difficult/low-margin to run as a business, but I think the co-op model you mentioned has potential. [0]: https://ne
28.
by supermdguy 4mo ago
It's interesting because their last model series (Phi) was based around the thesis that high-quality synthetic data is better than a large pre-training corpus.
29.
by supermdguy 4mo ago
I think the conduit exception still applies for analog faxes. Which makes no sense, since tapping a fax line is probably way easier than compromising a data center.
30.
by supermdguy 5mo ago
Apparently from a third party seller in New York....and tastes really bad. I was surprised steak could be safely mailed in such normal looking packaging! https://www.delish.com/food/a70539084/temu-meat-review/

More ›