Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
lopatin
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
12 ms
·
151.
▲
by
lopatin
3y ago
I should clarify that she would be able to pick up MySQL or Postgres without too much trouble.
152.
▲
by
lopatin
3y ago
That's a good idea. She knows finance/markets stuff too. I guess I thought banks had kind of cut throat cultures, but I agree that it's not necessarily like that in all departments. We'll for sure check out analyst posit
153.
▲
by
lopatin
3y ago
lol, maybe :)
154.
▲
Ask HN: How can my mom downshift as a SQL developer?
115 points
by
lopatin
3y ago
|
122 comments
155.
▲
by
lopatin
3y ago
Oh I read about the about the DK effect a while ago. I'm pretty much an expert in Psychology now, AMA.
156.
▲
by
lopatin
3y ago
I think the developer is justified in his self-described rant. He has gone through other channels and engaged with the project leaders directly. He has waited years for the bugs to be addressed. Microsoft is in the wrong here for sellin
157.
▲
by
lopatin
3y ago
I asked it to write an elasticsearch query for me and it worked, thanks!
158.
▲
by
lopatin
3y ago
It's usually up. No way this unrelated.
159.
▲
by
lopatin
3y ago
I used it for the first time today too, for the same reason. It was slower and much worse at coding. I was just asking it for SQL aggregation queries and it just ignored some of my requirements for the query.
160.
▲
by
lopatin
3y ago
Does this extend to politeness? If, in the training set, people are probably more likely to be helpful to a polite question, does that mean that ChatGPT will be more helpful if I ask the question politely? My intuition is yes, but I wonder
161.
▲
by
lopatin
3y ago
Sure, we can say that advertising in general is psychological manipulation. But that isn't what ajkjk was saying. They were ok with advertising, but not ok with targeted ads. > they should show fewer ads that don't fuck with ou
162.
▲
by
lopatin
3y ago
What makes a targeted ad the same as psychological manipulation?
163.
▲
by
lopatin
3y ago
This review is so long that it is actually a good substitute for reading the actual book. It covers the main points, and skips over hours of fluff, such as Lewis's constant personality analysis of SBF. Also I think it importantly touch
164.
▲
by
lopatin
3y ago
I believe that IntelliJ can do this for remote servers. It installs an IntelliJ agent on the server and it’s like remote development, your real IDE becomes a lightweight client of it.
165.
▲
by
lopatin
3y ago
It’s too late. People already call it hallucination and confabulating sounds weird. The idea that you think you can reverse the trend with a HN post is interesting though.
166.
▲
by
lopatin
3y ago
Ignoring it completely is a viable strategy. A lot of 12 factor sounds like common sense to people who regularly ship stuff. For example, if I fail to put the config in the environment, the risk that introduces will be immediately obvious i
167.
▲
by
lopatin
3y ago
I wouldn't be against some regulations that require safety critical AI models to be fully explainable. Or at least to some extent that "look what apparently ChatGPT can do" wouldn't qualify for.
168.
▲
by
lopatin
3y ago
I see, I think they're just broken for me in general.
169.
▲
by
lopatin
3y ago
These github.io pages always go down once they hit the front page.
170.
▲
by
lopatin
3y ago
I've wondered about this. If it's trained on human responses, and humans are more likely to be helpful to a polite question, then shouldn't ChatGPT also be more inclined to give helpful replies if I say please and thank you?
171.
▲
by
lopatin
3y ago
Devs who know they need an event queue but are not technical enough to set one up in SQS sounds like a very niche market to me.
172.
▲
by
lopatin
3y ago
> Leveraging public cloud infrastructure has traditionally demanded sifting through layers of boilerplate code and intricate configurations, often monopolizing a significant chunk of the developer’s time and energy. I don't buy this
173.
▲
by
lopatin
3y ago
I perhaps misunderstood that they get a small life time cut of my salary. You may be right that they get a bigger percentage of yearly, but as a lump sum, just of the first year.
174.
▲
by
lopatin
3y ago
I believe that my firm only does Chicago, but there are similar firms that contacted me that were for NY based companies.
175.
▲
by
lopatin
3y ago
It was this way, my yearly comp is whatever they offered me. What my recruiter makes is not my business, and I just assumed that they took a cut. I misspoke when I said "Maybe I did understand", I meant, "Maybe I did MISunder
176.
▲
by
lopatin
3y ago
Maybe I did understand. But I was under the impression that the company gets paid on a percentage of the salary the negotiated for me, and that she get's some of that too.
177.
▲
by
lopatin
3y ago
Thanks for your input. I did think that being exclusive was not great, but with the amount of interviews I was getting at companies that I was interested in, I was like "this should be work". To bring this down to numbers, years a
178.
▲
by
lopatin
3y ago
I'm talking about the ones that are employed by some recruiting firm. Their incentive to help you is that they get a small cut of your salary, which is paid by the company who hires you. And while, yes, you can argue that it is still i
179.
▲
Ask HN: Do You Use Recruiters?
11 points
by
lopatin
3y ago
|
19 comments
180.
▲
by
lopatin
3y ago
I contacted on LinkedIn by a recruiter, who I saw there were mutual acquaintances with and she even was my dads recruiter for a while. She did her work and now gets a percentage of my salary. That was an arrangement between her and my compa
More ›