Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
ubutler
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
7 ms
·
31.
▲
by
ubutler
1y ago
In my experience, maintaining a very popular software library, supporting open source, and blogging have absolutely all contributed to my success, and, additionally, as someone who is now a founder seeking like-minded, highly skilled engine
32.
▲
by
ubutler
1y ago
Further to @dust42, BERT is an encoder, GPT is a decoder, and T5 is an encoder-decoder. Encoder-decoders are not in vogue. Encoders are favored for classification, extraction (eg, NER and extractive QA) and information retrieval. Decoders a
33.
▲
We evaluated 112 terms of service: Nintendo had the worst
(isaacus.com)
3 points
by
ubutler
1y ago
|
0 comments
34.
▲
by
ubutler
1y ago
> Actually it reflects the idea of Unicode code points correctly. They are meant to represent graphs, not semantics. Why do we then have lots of invisible characters that are intended essentially as semantic markers (eg, zero-width space
35.
▲
Judge rules that police can bar high IQ scores (1999)
(nytimes.com)
10 points
by
ubutler
1y ago
|
1 comments
36.
▲
by
ubutler
2y ago
Hey HN, Over the past couple months, we, a team of Aussie legal and AI experts, have been working on building a new type of legal AI company — a company that, instead of trying to automate legal jobs, is trying to automate legal tasks. We w
37.
▲
Show HN: We built the first legal AI API
(isaacus.com)
1 points
by
ubutler
2y ago
|
1 comments
38.
▲
Show HN: The first legal AI API
(docs.isaacus.com)
5 points
by
ubutler
2y ago
|
0 comments
39.
▲
by
ubutler
2y ago
Kagi Professional is $10/m and comes with unlimited searches.
40.
▲
Kanon: the best legal AI classifier
(isaacus.com)
2 points
by
ubutler
2y ago
|
0 comments
41.
▲
Introducing Kanon: the world's best legal AI classifier
(isaacus.com)
4 points
by
ubutler
2y ago
|
0 comments
42.
▲
Ask HN: What's the easiest way to setup a scalable API developer platform?
2 points
by
ubutler
2y ago
|
0 comments
43.
▲
by
ubutler
2y ago
+1 It’s almost as indispensable as tqdm for a data scientist at least.
44.
▲
by
ubutler
2y ago
MessagePack can encode rows as well and then you just need to manage linking the keys during deserialization. In fact, it can encode arbitrary binary without needing base64 like JSON.
45.
▲
by
ubutler
2y ago
Although MessagePack is definitely not a drop-in replacement for JSON, it is certainly extremely useful. Unlike JSON, you can’t just open a MessagePack file in Notepad or vim and have it make sense. It’s often not human readable. So using
46.
▲
by
ubutler
2y ago
Nice. You can do the same in Python like so and surprisingly it only takes a couple seconds to a minute to get the print out, no need for importing any special libraries (just `sys` to enable printing large numbers). ```python import sys sy
47.
▲
by
ubutler
2y ago
I had already started using Kagi to create bangs that run searches like “site:reddit.com %s” but glad to see this made even easier!
48.
▲
The simhash patent has expired and is now free to use
(patents.google.com)
12 points
by
ubutler
2y ago
|
1 comments
49.
▲
by
ubutler
2y ago
Simhash is an extremely fast and simple algorithm for detecting near duplicate text at scale which makes it particularly useful for deduplicating AI training datasets.
50.
▲
by
ubutler
2y ago
Is it “this couple reign” or “this couple reigns”? The former just feels wrong despite it being the title.
51.
▲
Crafting a 13KB Game: The Story of Space Huggers
(frankforce.com)
4 points
by
ubutler
2y ago
|
0 comments
52.
▲
by
ubutler
2y ago
* without already labelled training data (assuming you're referring to causal LLMs). If you have labelled training data (or semi-labelled), BERT takes the cake, both in terms of accuracy and efficiency. In fact, you can have luck with
53.
▲
How to classify long documents without breaking the bank
(umarbutler.com)
1 points
by
ubutler
2y ago
|
0 comments
54.
▲
ChatGPT is a blurry JPEG of the web (2023)
(newyorker.com)
3 points
by
ubutler
2y ago
|
0 comments
55.
▲
by
ubutler
2y ago
I misinterpreted the title as Apple approving development of iOS apps on PC. That would’ve been an exciting development.
56.
▲
TuneMyMusic – transfer entire music libraries between streaming platforms
(tunemymusic.com)
2 points
by
ubutler
2y ago
|
0 comments
57.
▲
by
ubutler
2y ago
> and indeed, much of what used to be fertile farmland in that part of the world is now desert from over-farming Do you have a source for this?
58.
▲
The complete Code of Hammurabi (~1750 BC)
(ehammurabi.org)
84 points
by
ubutler
2y ago
|
55 comments
59.
▲
by
ubutler
2y ago
Probably because the title of the article uses 402,000,000 Mbps and I think there’s a rule(?) on HN about not misrepresenting titles or rewording them too much, though I’d wager 402 Tbps is fair alteration.
60.
▲
Microsoft shelves underwater data center – had less failures than those on land
(tomshardware.com)
4 points
by
ubutler
2y ago
|
0 comments
More ›