anax32

6 ms·

1.
by anax32 27d ago
Love that page styling.
2.
by anax32 1mo ago
I'm building a simple model inference platform: https://marigold.run Seems everyone is building an inference platform lately, but this one is mine! And I'm enjoying it...!
3.
by anax32 2mo ago
With new tools (i.e., LLMs) it feels like software engineering is moving to an assembly process. The best engineers will have an awareness of overall problem, architecture, how pieces should fit together and what changes when the fits are m
4.
by anax32 2mo ago
yes, that's what I thought. "You got a loicense for that PSU?" is my biggest fear!
5.
by anax32 3mo ago
A ban on desktop PC components over a certain power threshold (300W) or PSUs would definitely affect local AI, and Europe is not a big enough market, nor has it's own internal supply chains to offer alternatives. This has been done in
6.
by anax32 3mo ago
And it is focussed on social networks, which require an email address, which usually implies a device. But instead of inserting controls around email addresses (as with paid services) or devices (as with contraband), the requirement is push
7.
by anax32 3mo ago
The US "product machine" is so strong. They really know how to do frictionless signup and vendor lock-in on the corporate side.
8.
by anax32 3mo ago
Can anyone explain how this ultrasound can see through the skull? I've worked on ultrasound devices and data, the shadows from bone, and distortions caused by tissue types were very difficult. If this device can deal with those distort
9.
by anax32 3mo ago
Yes, thank you. That's exactly where I am, and trying to gather some knowledge. The power draw from the wall is especially important, because a spike across multiple devices at the same time can cause issues which are really difficult
10.
Ask HN: How is GPU power draw measured at scale?
6 points by anax32 3mo ago|2 comments
11.
by anax32 3mo ago
Open weight and local hosting is far, far cheaper. In every respect. Even support is cheaper, over time. However, it's difficult to sell this to businesses who want contracts and KPIs, not staff and commitments. Regulated industries wi
12.
by anax32 3mo ago
I've just made a milestone on my project, moving away from AWS (budget) to self-hosted and the local models are so much faster than in the past. Beyond LLMs, having embeddings, image, video, audio gen available is crazy. Running locall
13.
by anax32 3mo ago
I'm not getting any work done today thank you!
14.
by anax32 3mo ago
* You will have logs Always gets me
15.
by anax32 3mo ago
This was a big moment for me, but I now believe it's fictional. Thanks Ballard
16.
by anax32 4mo ago
This stuff is better than TV
17.
by anax32 4mo ago
HTTP content negotiation was a good idea which decouples content from form, but only as far as format selection. Generative models are able to transform content between media types, and it feels like the original intention can be completed
18.
by anax32 4mo ago
I've been thinking of LLM prompting as execution, which makes context building a compile step -- different build systems, different output quality. Chunking is the simplest compiler; artefact construction that infers schema and resolve
19.
Show HN: JSON-logic-path – JSON logic with jsonpath multi-value resolution(github.com)
2 points by anax32 5mo ago|0 comments
20.
by anax32 5mo ago
an interesting aside: the block rewards for each epoch are the percentage issuance of that epoch.
21.
Content vs. Curation(bayinformationsystems.substack.com)
2 points by anax32 5mo ago|0 comments
22.
Digital Archaeology: where does Bitcoin's 21M come from?(bayinformationsystems.substack.com)
2 points by anax32 5mo ago|1 comments
23.
AI development tools shouldn't be faster horses(bayinformationsystems.substack.com)
1 points by anax32 7mo ago|0 comments