Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
aakresearch
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
5 ms
·
31.
▲
by
aakresearch
4mo ago
- A little script that is best described as "symbolic submodule" workflow enabler for git: https://github.com/fontmaniac/symgit . As in "symbolic link" vs "hard link" - where hash tying hos
32.
▲
by
aakresearch
5mo ago
> I fear that workflow is just chat, search, and being a rubber duck for my thoughts This is exactly what I settled upon after my own trying really hard. It is liberating, I have no fear at all!
33.
▲
by
aakresearch
5mo ago
Oops... I am deeply sorry, thank you for the heads up! It seems I've myself committed a cardinal sin that I am usually quick to point in others - rushing to reply without comprehending the full message. (Meta-oops: I realized how LLM-i
34.
▲
by
aakresearch
5mo ago
>>At the same time, do you really want every conversation you have with your doctor recorded >Yes. This is what medical records are. No. Medical records are limited extracts from conversations , which is your doctor and only
35.
▲
by
aakresearch
5mo ago
I may have different perception facilities in the brain, but to me the images in the article are horrifying. Such bad, Frankenstein-level, criminal butchering of absolutely everything!
36.
▲
by
aakresearch
5mo ago
I am by no means an expert, but I'd like to offer my mental model - up to you to decide if it is solid or not, but it works for me. I think the core intuition is that, like with any other "rasterized" system with finite memor
37.
▲
by
aakresearch
5mo ago
It seems that author unironically advises to write your commit messages like this: "Restructured Claude’s module architecture, rejected initial state management approach, rewrote error handling from scratch", to have a chance at d
38.
▲
by
aakresearch
7mo ago
To my understanding, LLM, by design, is unable to encode negation semantics. Neither negation "operation", nor any other "subtractive" operations are computable in LLM machinery. Thinking out loud, in your example the &q
39.
▲
by
aakresearch
7mo ago
Agree. Your post mirrors surprisingly well with mine of last year [0] - did you, by chance, happened to come across it? I said "you’d likely get approximately the same quality of candidates from a randomly selected 50 out of 5,000"
40.
▲
by
aakresearch
7mo ago
I am in agreement with you, but regret that you missed an opportunity to swap two paragraphs around and purposefully mislabel them (i.e. the LLM-generated as your own, and vice versa). I'd be very curious if audience here would success
41.
▲
by
aakresearch
7mo ago
I think you are mistaken. I remember landing upon that website some time before 2023 and being greatly impressed by that very 3D parallax effect. To my knowledge there were no AI tools available in 2022 capable of producing effects like tha
42.
▲
by
aakresearch
8mo ago
Fictional artist Feofan Kopytto, who was immortalized in "The Little Golden Calf" [0] used oats and other cereals for his paintings. On the back of the book authors' talent it became customary to refer to his artistic endeavo
43.
▲
by
aakresearch
8mo ago
I am in agreement with many commenters here ( https://news.ycombinator.com/item?id=47158240 , https://news.ycombinator.com/item?id=47158573 and others) that this article is a clear illustration of failure on
44.
▲
by
aakresearch
8mo ago
Such approach is not even new, and has been conceptualized centuries ago! [1] [1] https://en.wikipedia.org/wiki/Stone_Soup
45.
▲
by
aakresearch
8mo ago
This is a very useful take, thank you. Really helped me to adjust my mental model without "antropomorphising" the machinery. Upvoted. If I may, I would re-phrase/expand the last sentence of yours in a way that makes it even m
46.
▲
by
aakresearch
8mo ago
Upvoted, as it basically 99% matches my own thinking. Very well said. But I, personally, would not predict a breakthrough in this direction in the next 2-5 years, as there is no pathway from current LLM tech to "true reasoning".
47.
▲
by
aakresearch
11mo ago
Wow, sounds so familiar! I've once had to argue precisely against this very conclusion - "you saved us once in emergency, now you're bound to do it again". Wrote to my management: "It is, by all means, great when a
48.
▲
by
aakresearch
11mo ago
Now do LinkedIn Simulator!
49.
▲
by
aakresearch
1y ago
Like a proverbial broken clock which shows correct time twice per 24 hours, AI may "show correct time" for 99% of prompts, but doesn't deserve any more trust.
50.
▲
by
aakresearch
1y ago
> Pat yourself on the back and mutter “you need to be skilled and lucky to work here” It would be absolutely amazing if employers and recruiters finally were doing exactly this. We are in this dead end precisely because everyone is under
51.
▲
by
aakresearch
1y ago
Your take is very sensible and I agree with it 100%, but the reality is that (by my assessment) it is absolutely not present in the wall of ATS filters one's job application is up against. I've sent hundred of CV/cover letter
52.
▲
by
aakresearch
1y ago
Disclaimer: I've been navigating the job market for almost a year now, and have practically given up. An acquaintance of mine—he was the owner's/CEO's deputy at a place I worked—now runs a team-coaching-turned-recruitmen
53.
▲
by
aakresearch
2y ago
Brilliantly put! Upvoted and "favorited". I would also add meticulous attention to documenting requirements and decisions taken along the development process, especially where compromises were made. All the "why's",
54.
▲
by
aakresearch
2y ago
Doesn't sound absurd at all to me. Divorce is bad enough so the child can perceive it as having the same "finality" as death, with additional burden of abandonment, disowning and guilt without any chance of reconciliation. Ka
55.
▲
by
aakresearch
2y ago
How many would you consider un-acceptable to the point of introducing 2020-level of insane restrictions?
56.
▲
by
aakresearch
2y ago
I feel it is fruitless to continue this conversation as it doesn't seem that we may find agreement, but if you insist.. (In the following paragraph "you" is mostly figurative, not personal) Yes, I let a lot of anger and emoti
57.
▲
by
aakresearch
2y ago
It has nothing at all to do with treatment being unequal. I only form my opinion from the current Wikipedia article, and this is how it looks to me. - Mary Mallon, herself, was just as innocent, yet she was treated as a criminal, which is u
58.
▲
by
aakresearch
2y ago
Eloquent primer on "cogent argument" which is totally not "rambling stream of anger and emotion". Perfect basis for a healthy, respectful debate. I'm really sorry I did bite on this provocation, but hope your post s
59.
▲
by
aakresearch
2y ago
Case in point.
60.
▲
by
aakresearch
2y ago
Yeah, it was a mistake to expect anything beyond thought-stopping cliches that were delivered by you. In hindsight it was a mistake to post at all. I don't think you've read my post anyways, or you've processed it on a level
More ›