Hacker News new | past | comments | ask | show | jobs | submit | jkukul's comments login

I use iOS's built in Screen Time settings. For "bad" apps (Reddit, TikTok, etc) and "bad" websites ("hackernews", etc) I set a daily time limit of, let's say, 15 minutes.

I configure a random password for Screen Time so that it's a real hassle to circumvent the daily limit when I get over it.


Yes, you can pre-fill the assistant's response with "```json {" or even "{" and that should increase the likelihood of getting a proper JSON in the response, but it's still not guaranteed. It's not nearly reliable enough for a production use case, even on a bigger (8B) model.

I could recommend using ollama or VLLm inference servers. They support a `response_format="json"` parameter (by implementing grammars on top of the base model). It makes it reliable for a production use, but in my experience the quality of the response decreases slightly when a grammar is applied.


Grammars are best but if you read their comment they're apparently using ollama in a situation that doesn't support them.


I think in the era of LLMs good docs/FAQ are of an even greater value.

You can write a support bot that sends a user's question + docs/FAQ to an LLM to automatically deal with the basic questions and only involve a human in the loop once a question goes beyond what's in the docs.


For a side project?

Not even the big giants manage to create LLM bots that work.


A leet-code test would be much more standardized if candidates could solve it at home. Just send me a link to the quiz and let me solve it within a specified time frame.

I've done tests like this for some companies. It felt a lot fairer and more closely resembling the actual work environment than live leet-code interviews, with biased interviewer(s) and a stress factor that's not a part of the actual job.


As a hiring manager I HATE leet-code tests, and they do nothing to differentiate candidates, but a take home in the era where people run chatGPT beside the interview window, or have someone else do the interview for them? Not a chance. You are 100% correct that it is way more representative, but the prevalence of cheating is ridiculous.


I totally understand you, but want to offer a different perspective.

They will also be able to use ChatGPT on the job. And StackOverflow. And Google. If they know how to use tools available to solve a problem, that will benefit them on the job.

If you're testing them for what ChatGPT can already solve, then are the skills being tested worth anything, in this day and age?

Take-home LeetCode, even with cheating will still filter out a good chunk of candidates. Those who are not motivated enough or those who don't even know how to use the available help. You'll still be able to rank those who solved the task. You'll still see the produced code and be able to judge it.

Like other commenter points out, you can always follow up the take-home LeetCode. Usually, it becomes apparent really quickly if a candidate solved it on their own.


This does seem like a vexing problem, especially when interviews are conducted remotely.

I wonder if either of the following could be cost-effective:

(a) Fly the candidate to a company office, where their compute usage could be casually monitored by an employee.

(b) Use high-quality proctoring services that are nearby to the candidate. E.g., give them 1-2 days in a coworking space, and hire a proctor to verify that thy're not egregiously using tools like ChatGPT.

Or alternatively, would it suffice to just have a long conversation with the candidate about their solution? E.g. what design trade-offs they considered, or how might they adapt their solution to certain changes in the requirements.


Before Covid, on-sites were, well, on site, and flying the candidate in for a day was just accepted practice.


Take home is fine if you discuss it later in the interview. But also there should be some pre-screening to keep the number of interviewees reasonable.


The better/new interview question, thus, is "here is code that chatGPT generated for $PROBLEM", what's wrong with it.


In theory, yes. Like you’re saying we’re still not quite there yet.

In practice there are constraints that limit job markets geographically, e.g. time zone differences or legal obstacles to hire foreigners.


Thanks for the insightful comment.

I guess they can still conveniently schedule when the layoffs are announced and make it happen right before the 10b5-1 schedule?

All the previous layoffs announcements were followed by the stock rising up.


Exactly. The 10b5-1 schedule is a layer of misdirection, but the executives still have their hands firmly on the levers of price manipulation.


There is no way to know what effect a layoff will have on stock price in advance. This sale would have happened in the exact same way if the stock had tanked after the layoff.


> There is no way to know what effect a layoff will have on stock price in advance

Sure, but looking at precedent gives you info to make a good bet- the last 2 Spotify layoffs this year each caused their stock to jump.


Sure. It's possible he decided he wanted to use a layoff as a way to make a gamble on the stock price before a scheduled sale.

However there is no actual evidence of this, so let's be careful about breaking out the pitchforks. This sale would have happened even if the stock tanked 50%.


It's not just Spotify layoffs that have caused stock pumps this year. Plenty of other large tech companies have experienced the same pump due to layoffs. It's a pretty good gamble.

The 10b5-1 schedule is referenced often by executives as an alibi against insider trading. Let's just be realistic that it likely doesn't prevent anything of the sort.


It is. At best. A gamble. There are more than enough examples of companies doing layoffs and experiencing stock losses.

Let’s be realistic that this trade was a) always going to happen no matter the stock price, b) the CEO does not have a crystal ball, and c) there is no evidence of the CEO timing layoffs as a ploy for personal gain.

It’s easy to get caught up in conspiracies that we want to be true.

Imagine if the timing were juuuust a little different, what the headlines would be. "Spotify CEO cashes in on millions before announcing layoffs". Hmmm.


It's weird to deny that executives can comprehend the impact of their decisions on the price of their ticker, but ok.

> It’s easy to get caught up in conspiracies that we want to be true.

I'm not saying there even needs to be a conspiracy for this specific case. I'm pointing out that there is nothing in the law that would stop them from taking advantage of this timing. Also that previous layoffs have resulted in a bump in share price multiple times in a row, and what do you know, it happened again. Even if it was a gamble, how can you deny that they're able to time information drops to their personal advantage? Isn't that a problem in itself?


I haven't denied any such thing. I'm literally just saying it's a gamble.

Is it a problem? It seems unsolvable if so, unless we should ban CEOs from receiving stock compensation?


>I haven't denied any such thing. I'm literally just saying it's a gamble.

Well, you said it was "at best a gamble". I don't know how else I'm supposed to interpret that besides you saying there was "at best" a 50/50 shot of either outcome happening from the perspective of the ones that announced it. I'm going to assume this is not what you meant.

Anyway, since it's established that we're in agreement (maybe? It's not easy to figure out your point of view) that executives are generally able to know with some degree of certainty >50% that their actions will create a specific effect on their stock price. And we also can agree that it is their decision when to announce and execute these actions. It's clearly and obviously a problem that they also are aware of when the stock that they hold in large quantities will be bought and sold.

> Is it a problem? It seems unsolvable if so, unless we should ban CEOs from receiving stock compensation?

If we can agree on the previous bit, then I have no idea why we should throw up our hands and be like "well I guess this is an impossible problem to solve". It's really not. Just be more restrictive with the windows in which executives are allowed to sell. Better yet spread their entire buy/sell order over the course of the year. Or even better still, don't allow them to sell stock AT ALL until they are not in charge or aware of high level decisions at the company.

CEO pay is on average 399x their average employee(https://www.epi.org/publication/ceo-pay-in-2021/). Then, we also allow them to pump their portfolio on a whim at the expense of their employees. I would say personally I don't really care if their compensation suffers, since they've probably already 10x'd my lifetime earnings in their last year of work, and I'd happily switch places with them financially if they're so hard done by.


> Just be more restrictive with the windows in which executives are allowed to sell.

I feel like I'm on crazy pills.

Executives. Do. Not. Have. Windows. To. Sell. I don't know how many times I have to repeat this.

They have exact dates with exact numbers of shares on which sales will execute with exactly zero input from them. It's literally the most restrictive thing you can imagine, short of not allowing them to engage in stock sales at all.

https://www.raymondjames.com/-/media/rj/advisor-sites/sites/...

Here is what you seem to think happened: Daniel Ek fired a bunch of people during an open trading window, saw that the stock price went up, and sold a bunch of shares.

Here is what actually happened: Daniel Ek put together a schedule of specific stock sales at the beginning of 2023 that executed automatically on his behalf throughout the year. At some point during the year, he decided to fire a bunch of people.

> Better yet spread their entire buy/sell order over the course of the year.

They do. It's on a schedule. Over the course of the year.

> Or even better still, don't allow them to sell stock AT ALL until they are not in charge or aware of high level decisions at the company.

This is effectively banning equity pay for executives. Unless you think we should make C level executives somehow not in charge of high level decisions.

> CEO pay is on average 399x their average employee

An entirely different discussion.


> I feel like I'm on crazy pills.

Could be.

> Here is what you seem to think happened: Daniel Ek fired a bunch of people during an open trading window, saw that the stock price went up, and sold a bunch of shares.

Nope.

> Executives. Do. Not. Have. Windows. To. Sell. I don't know how many times I have to repeat this.

You can stop whenever? Obviously I know this. Windows are just further restrictions to add to the schedule. I'm certainly not an expert, but I dunno why the concept of combining restrictions is so hard to comprehend.

You're on your own for the rest of the misinterpretations though. I think those ones are easier if you think a little harder about it.


Because the existing restriction, is already, more restrictive, than windows. How is this so hard.

Like honestly. Let's say Daniel Ek's 10b5 says he will sell 10,000 shares on 1/1, 3/1, 6/1, and 9/1 of 2023.

What on earth would adding trading windows do to make that more restrictive?

You're asking for a 50mph speed limit when the car's max speed is 25. It's completely pointless and accomplishes nothing.


> I guess they can still conveniently schedule when the layoffs are announced and make it happen right before the 10b5-1 schedule?

This is still a gamble at best. Let's not go down a conspiracy rabbit hole here.

CEOs have their trading schedules set well in advance, they do not know how the market will respond to their actions ahead of time. They can at best make an educated guess.

But the idea that he scheduled this layoff specifically for personal gain is pretty unfounded.


There was a very interesting discussion recently on this topic. [1]

There are various hypothesis:

* changes to gut microbiome composition due to modern diets and overuse of antibiotics

* excessive cleanliness and hygiene. The "hygiene hypothesis" suggests that lack of exposure to diverse microbes leads to improperly trained immune systems

* increased exposure to chemicals, plastics, and pollutants that may disrupt immune function

* Cesarean sections and formula feeding rather than vaginal birth and breastfeeding

[1]: https://news.ycombinator.com/item?id=37195905


Exposure to cigarette smoke was also considered a candidate or reason, when I was younger. I haven't kept up with the thinking on or about this, but I believe it's still considered valid.

Remember, a few decades ago, a lot of the "Western" world smoked, including in public spaces and around kids. As I understand it, a lot of the rest of the world has caught up and still does so. And the Western world isn't exactly rid of the behavior, either.

P.S. I wonder whether anyone's studied possible multiple-generation, genetically-communicated knock-on effects of this.

P.P.S. "Pollen" is also a Racket language based blogging and authoring tool or environment. A bit sorry to see the name become overloaded.


Yes, that's the main point, the stronger concrete. If you make stronger concrete it will last longer and you thus will need less of it in the long run.

Concrete production is very bad for environment, so the less we need to make the better.


And Australia: know for insane coffee consumption (has to be freshly ground!) and appartments with “concrete cancer” although that may not be solved by this.


Or one layer down (= founders), let’s not forget Adam Neumann renting his properties to WeWork


$100k salary puts you in the top 5% of the society in Bulgaria. I wouldn't call that being "Ramen-fueled". I don't want people to get wrong ideas.

I agree with you that $235K is a small budget to build a serious software though.


It's actually a lot less than that. It's more like $50k. Just saying.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: