nancarrow
Re:Generative AI floods and dilutes the market for books
on Sunday August 02, 2026 @08:16AM (#49143799)
90% of crud filtering mechanisms are crud.
Re:System: Control your Mac from anywhere using natural language
on Tuesday January 06, 2026 @11:18AM (#46514308)
love how clearly vibecoded this is. the cloudflare worker architecture + the ascii diagram is a dead giveaway. nothing wrong with that, it's just really obvious.
the split architecture offers absolutely zero security benefits outside of not exposing a server process on your mac to the open internet (assuming you only let it connect to cloudflare) - it's just a convenient place to spin up a thin JS layer that calls model APIs and connects to your mac.
anyways i think this is a neat weekend vibecoding project but IMO it needs a lot more design thought to really be useful and not be a huge security issue.
Re:Google Gemini: Context Caching
on Thursday May 16, 2024 @04:05PM (#40382535)
"some" is doing a lot of lifting. # of tokens * # of layers * head dimension * # of heads * 2 (K+V vectors) * 4-16bits (depending on quantization)
Re:Correctness and composability bugs in the Julia ecosystem
on Wednesday May 18, 2022 @04:52PM (#31427498)
> How many Matlab users know things could be better?
not very many in my experience - matlab rots the brain
Re:The Mathematics of Artificial Intelligence (2022)
on Monday April 11, 2022 @12:57PM (#30991617)
the subclass of "Reproducing kernel Hilbert spaces" form the basis for a large class of ML algorithms (kernel methods)
Re:GenieFramework – Build web applications with Julia
on Wednesday April 06, 2022 @01:21PM (#30934982)
why? i like Julia but it's not a great general purpose language IMO.
Re:Ask HN: Who operates at scale without containers?
on Tuesday March 22, 2022 @02:21PM (#30769551)
the CUDA library versions are very coupled to the hosts' CUDA driver versions and the container system itself (Docker) needs special code to link the two
Re:How percentile approximation works and why it's more useful than av...
on Wednesday September 15, 2021 @09:47AM (#28538976)
It's more related to the law of large numbers than the CLT
Re:Ideas in statistics that have powered AI
on Wednesday July 07, 2021 @10:36PM (#27767965)
The generator implicitly models a joint probability of data by being a generative process that one can draw samples from. GAN training (at least under certain simplifying assumptions) minimizes the JS divergence between the generator distribution and the data distribution.