Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
overbring_labs
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
7 ms
·
1.
▲
Qwen3.8-27B: slower tokens, faster and better results
(overbring.com)
1 points
by
overbring_labs
1mo ago
|
0 comments
2.
▲
The exceptionalism of software is unwarranted
(overbring.com)
3 points
by
overbring_labs
1y ago
|
0 comments
3.
▲
Things I've realized about testing Elixir code
(overbring.com)
2 points
by
overbring_labs
1y ago
|
0 comments
4.
▲
I don't buy Macs anymore
(jasonsaidwhat.substack.com)
24 points
by
overbring_labs
1y ago
|
35 comments
5.
▲
Vibe coding leaves me with a very sour taste
(overbring.com)
7 points
by
overbring_labs
1y ago
|
0 comments
6.
▲
by
overbring_labs
1y ago
Well, it's in Rust now, thanks to your and someone else's comments! https://github.com/waseigo/disk_space
7.
▲
by
overbring_labs
1y ago
Truly, way better than before: https://github.com/waseigo/disk_space And it now runs on everything, across more Elixir and Erlang/OTP versions than before!
8.
▲
by
overbring_labs
1y ago
Thanks, one day I'll look into Zig too, because I've been impressed with what Tigerbeetle has achieved with it! For now, I'm all-in on Elixir.
9.
▲
by
overbring_labs
1y ago
Holy moly, you weren't kidding. Grok 4 is so much better. Thanks!
10.
▲
by
overbring_labs
1y ago
Given the amount of issues the code had when I ran splint on the C file, I agree. The question was for me whether I can get something working to get over the "speed bump" of lacking such a function for the API client I'm writ
11.
▲
by
overbring_labs
1y ago
Amen, thank you for noticing. The goal here was not to produce something of stellar quality, which is anyway out of the question as I don't have the skills/knowledge to evaluate anything other than "it returns the Elixir map
12.
▲
by
overbring_labs
1y ago
It's just that Grok 3 is faster than 4, so I've set it by default. But point taken, I'll try out the newer ones now that I'm converting it to Rust.
13.
▲
by
overbring_labs
1y ago
Because what difference would it make, given the bad quality of code? Also, is Claude Code free to use? The manual process has the upside that you get to see how the sausage is (badly) made. Otherwise, just YOLO it and put your trust in Gen
14.
▲
by
overbring_labs
1y ago
What is your definition of "a working project"? It does what it says on the tin (actually it probably does more, because splint throws some warnings...)
15.
▲
by
overbring_labs
1y ago
It's more of "yeah it worked, but I had to do a lot of hand-holding" and "it passes the tests but I cannot tell if the code has memory leaks". Actually, I can tell; I ran split on the C source and got things like th
16.
▲
by
overbring_labs
1y ago
I mean, you aren't wrong. I'm looking into converting it into Rust with Rustler right now.
17.
▲
I let LLMs write an Elixir NIF in C; it mostly worked
(overbring.com)
71 points
by
overbring_labs
1y ago
|
69 comments
18.
▲
My Dad's Last Days
(petros.blog)
4 points
by
overbring_labs
1y ago
|
1 comments
19.
▲
by
overbring_labs
2y ago
You wouldn't ask a rabbi about the New Testament or an imam about the Torah and expect unbiased responses. So why ask a CCP-influenced LLM about things you already know you won't get an unbiased answer to?
20.
▲
by
overbring_labs
2y ago
Stop following the news? My recipe for success for 10 years now and counting. The news is meant to elicit views, clicks, and emotional reactions. Why would you subject yourself to that on a regular basis, or at all?