alexmolas

6 ms·

31.
Hand-Drawn QR Codes(sethmlarson.dev)
7 points by alexmolas 1y ago|0 comments
32.
Is There a Half-Life for the Success Rates of AI Agents?(tobyord.com)
5 points by alexmolas 1y ago|0 comments
33.
Library of Babel(libraryofbabel.info)
2 points by alexmolas 1y ago|0 comments
34.
by alexmolas 1y ago
I used to work on a logistics company and we had to map latitude and longitude to specific directions. One of the first things I learnt was to avoid storing 6 decimal precision coordinates. Also, this XKCD was shared a lot https:/&#x2
35.
Unpredictability and Undecidability in Dynamical Systems [pdf](gwern.net)
5 points by alexmolas 1y ago|0 comments
36.
A Consequentialist Critique of Binary Classification Evaluation Practices(arxiv.org)
3 points by alexmolas 1y ago|0 comments
37.
How to Provide Feedback on Documents(lethain.com)
2 points by alexmolas 1y ago|0 comments
38.
by alexmolas 1y ago
This is textbook survivorship bias. Out of 133 electable cardinals, someone was bound to guess Robert Prevost. If they were wrong, no one would remember. You could probably find 132 others who guessed wrong.
39.
by alexmolas 1y ago
how does it compare against mypy? is it much faster?
40.
docker2exe: Convert a Docker image to an executable(github.com)
142 points by alexmolas 1y ago|90 comments
41.
Semantic unit testing: test code without executing it(alexmolas.com)
75 points by alexmolas 1y ago|71 comments
42.
by alexmolas 1y ago
> Religion needs science to keep it away from superstition and keep it close to reality, to protect it from creationism, which at the end of the day is a kind of paganism – it's turning God into a nature god I love this quote!
43.
Vatican Observatory(vaticanobservatory.org)
126 points by alexmolas 1y ago|53 comments
44.
by alexmolas 1y ago
There are two examples in the repo, one with CoRT and another one without. And the one without it it's much better than the one that uses it. Weird choice of examples...
45.
by alexmolas 1y ago
The astral team (behind uv and ruff) is already working on a type checker in Rust, and given the quality of their existing tools, I'm inclined to wait and see what they release. Pyrefly looks interesting, but from the repo it seems pre
46.
Black Hole Information from Non-Vacuum Localised Quantum States(arxiv.org)
1 points by alexmolas 1y ago|0 comments
47.
DeepWiki(deepwiki.com)
2 points by alexmolas 1y ago|1 comments
48.
Show HN: I built a package for Semantic Unit Testing(alexmolas.com)
2 points by alexmolas 1y ago|2 comments
49.
GPT Image 1: image generation model(platform.openai.com)
2 points by alexmolas 1y ago|0 comments
50.
Llama 4 Smells Bad(fastml.com)
41 points by alexmolas 1y ago|26 comments
51.
Testament of Holy Father Francis(vatican.va)
3 points by alexmolas 1y ago|0 comments
52.
Semantic Unit Testing(alexmolas.com)
4 points by alexmolas 1y ago|1 comments
53.
by alexmolas 1y ago
is this the same as error propagation? I used to do a lot of that during my physics degree
54.
Optimizing PyTorch Docker images: how to cut size by 60% – Marton Veges(mveg.es)
2 points by alexmolas 2y ago|0 comments
55.
Smartfunc: Turn Docstrings into LLM-Functions(github.com)
70 points by alexmolas 2y ago|25 comments
56.
Symmetric Math Riddles(alexmolas.com)
1 points by alexmolas 2y ago|0 comments
57.
Three Symmetric Math Riddles(alexmolas.com)
1 points by alexmolas 2y ago|0 comments
58.
by alexmolas 2y ago
I guess the biggest limitation of this approach is that the max output length is fixed before generation starts. Unlike autoregressive LLM, which can keep generating forever.
59.
Vector Search with DuckDB(huggingface.co)
17 points by alexmolas 2y ago|0 comments
60.
Remote work sucks, but that doesn't mean it's over(revenuecat.com)
1 points by alexmolas 2y ago|1 comments

More ›