Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
paletov
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
4 ms
·
1.
▲
From RMSProp to AdamW: The Optimizer Evolution Story
(naskovai.substack.com)
2 points
by
paletov
1y ago
|
0 comments
2.
▲
The Sandwich Framework for Understanding Linear Algebra
(substack.com)
2 points
by
paletov
1y ago
|
0 comments
3.
▲
by
paletov
2y ago
Please note that this is currently optimized for parametric modeling of functional components, not organic shapes.
4.
▲
by
paletov
2y ago
Thanks! It’s a mix of some off-the-shelf models with custom training for CAD tasks.
5.
▲
by
paletov
2y ago
Thanks for the interest! Yeah, it does use OpenSCAD under the hood. The parameter names are something I'm still tweaking—glad to hear they worked well in the headphone hanger example!
6.
▲
Show HN: CAD with AI
(cadwithai.com)
8 points
by
paletov
2y ago
|
7 comments
7.
▲
by
paletov
3y ago
Currently it returns: grep -r --exclude-dir=.venv "class" .
8.
▲
by
paletov
3y ago
My understanding is that ShellGPT aims to be a complete OS assistant. It's similar to Open Interpreter ( https://github.com/KillianLucas/open-interpreter ). Shelly is a mini tool at the moment that only generates an
9.
▲
by
paletov
3y ago
It opens the command in nano/vim in the terminal and lets you edit it. Saving executes the final version. Currently, the user has to be experienced enough to know if a command is safe to be executed. However, you are right, and I'
10.
▲
by
paletov
3y ago
It's too early for cowboy mode. The LLM isn't reliable enough to let it run commands without approving them first. As others noted, "try" could be used to make the commands testable, so I'll look into it this weeken
11.
▲
by
paletov
3y ago
Yeah, that would be bad. That's why I added an intermediate step. It opens the command in an editor like when you run "git commit", so you can edit it. Saving executes the final version.
12.
▲
Show HN: Shelly: Write Terminal Commands in English
(github.com)
35 points
by
paletov
3y ago
|
54 comments