Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
aledalgrande
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
31.
▲
by
aledalgrande
1y ago
I've seen this comparison a few times already, but IMHO it's totally wrong. A compiler translates _what you have already implemented_ into another computer runnable language. There is an actual grammar that defines the rules. It d
32.
▲
by
aledalgrande
1y ago
Or a good commit/PR description?
33.
▲
by
aledalgrande
1y ago
Not just the SaaS community, unfortunately.
34.
▲
by
aledalgrande
1y ago
On a sidenote, I tried to get Codex + O3 to make an existing sidebar toggable with Tailwind CSS and it made an abomination full of bugs. This is a classic "boilerplate" task I'd expect it to be able to do. Not sure if I'
35.
▲
by
aledalgrande
2y ago
Good for you Ilya, best of luck with your company!
36.
▲
by
aledalgrande
2y ago
The thinking tokens (even just 1024) make a massive difference in real world tasks with 3.7 in my experience
37.
▲
by
aledalgrande
2y ago
LOOOOL you have my upvote (I did use Spring, once, ages ago, and we deployed the app to a local Tomcat server in the office...)
38.
▲
by
aledalgrande
2y ago
I know the language earned you the downvotes (please be kind), but the author of the article is ex Google and ex AWS, I too would expect some better infra in place (caching?) and certainly not Vercel.
39.
▲
by
aledalgrande
2y ago
> understanding the context of a conversation to adjust the tools dynamically This just sounds like a small fine tuned model that knows hundreds of MCP tools and chooses the right ones for the current conversation.
40.
▲
by
aledalgrande
2y ago
Seems like we're rediscovering/reinventing the same thing over and over, this is very close to Memory Bank from Cline https://docs.cline.bot/improving-your-prompting-skills/cline... I have seen a number of ar
41.
▲
by
aledalgrande
2y ago
You'd be happy to know we measure the size of horses in hands
42.
▲
by
aledalgrande
2y ago
> Over 43% of MCP server implementations tested by Equixly had unsafe shell calls. How can we fall into this _every single time_.
43.
▲
by
aledalgrande
2y ago
ROTFL at the abstract
44.
▲
by
aledalgrande
2y ago
They would become a senior developer? lol ;)
45.
▲
by
aledalgrande
2y ago
> Unless these tools can be run locally independent of a service provider, we're just trading one boss for another. Not only that, we have to be careful about all the integrations being built around it. Thankfully the MCP standard i
46.
▲
by
aledalgrande
2y ago
Am I the only one that finds these generation-related articles really stupid? Gen <put letter here> meltdown. There are stupid people in every generation. There are smart people in every generation. If the writer was in the second gro
47.
▲
by
aledalgrande
2y ago
The personal project linked in the article is so beautifully done. A joy to interact with
48.
▲
by
aledalgrande
2y ago
Oh man 3dfx and Matrox, so much nostalgia. I didn't have the money for any of those so they will always stay a legend for me. I think my dad's 586 had an S3 on it. In those times, the only way to find games walkthroughs and cheat
49.
▲
by
aledalgrande
2y ago
In my experience it's only (and exactly) leaders without tech nor people skills that do this. Have experienced both good and bad. A world of difference.
50.
▲
by
aledalgrande
2y ago
Figma is much more than that and lots of people use it daily
51.
▲
by
aledalgrande
2y ago
> dumb middle management that seem to always be leading software projects despite having no technical background Or you change to a company with capable & technical middle management :) It always blows my mind when people say "m
52.
▲
by
aledalgrande
2y ago
I had the same thought. Why do we need to keep coming up with weird names/acronyms/portmanteau
53.
▲
by
aledalgrande
2y ago
Same with Cline
54.
▲
by
aledalgrande
2y ago
Could you share how you track AI vs human LoC?
55.
▲
by
aledalgrande
2y ago
"And that screen resolution? Did you measure it in maple syrup units?" Typical roast of a Canadian.
56.
▲
by
aledalgrande
2y ago
https://ollama.com/library/qwq
57.
▲
by
aledalgrande
2y ago
I agree having worked directly with it (long personal project). Photogrammetry is so error prone (get a sign wrong in your sensor fusion/SLAM code and you're cooked for 2 weeks) and I got into it right in the last years before mod
58.
▲
by
aledalgrande
2y ago
Some visuals https://interestingengineering.com/science/project-orion-the... PDF in this article is a 404
59.
▲
by
aledalgrande
2y ago
50? Think 50000.
60.
▲
by
aledalgrande
2y ago
I was not a PhD, but I studied both around that period in university and can confirm that neural networks were only seen as a kind of theoretical plaything. All the libraries were pushing SVMs.
More ›