schneems

The pay question is valid, but not appropriate for a technical interview. If someone asked me I wouldn’t even know the answer. That would be a question for the recruiter or possible engineering manager.

I think it’s 100% okay to ask about pay in an interview but not okay if it’s the only thing you ask about.

For me: I care about the day-to-day of who I’m working with and what that dynamic is like (in addition to money and benefits).

Oh for sure. Ask questions, just not annoying whiteboard questions. Also I hate this “ask us anything” part of the interview. It’s so performative.

It should be rephrased as “the jeopardy round” since it’s still about the candidate, but phrased backwards. And it’s not a time for REAL questions, it’s a time to show you’re smart and attentive but not TOO smart, you want the interviewer to feel good about themselves so they can feel good about you.

> what my response would be

I don’t ask candidates to do anything I wouldn’t put up with. It would be unusual but I would be game (if they were serious). Fundamentally that’s what my fantasy is about: a world where interviewer and interviewee have mutual respect for each other.

In the recent past I’ve asked candidates to walk me through code they’ve written. I’m super happy to reciprocate for 15 min and I think the candidate (if they’re working with me directly) would get a lot out of it.

It’s a trick question. The number is a distraction. If they don’t ask the model year of the VW Beatle they’re clearly not detail oriented and can’t be trusted. /s

I’ve never actually done this. But I’ve fantasized about preparing several interview questions for the company I’m interviewing at. They forget that interviews are a two way street.

If I like them (and the process was bearable), I would ask nothing. If I’m mildly annoyed, something “simple” yet patronizing like fizbuzz. If I’m REALLY annoyed then something wildly specific and pedantic.

Interviewer: “do you have any questions for me?”

Why yes, a chicken, fox and sack of flour need to cross a doubly linked list, how would you flip the list inside out from the middle while counting the number of pingpongballs that can fit into 747 VW Beatles.

I learned to love rustfmt but there’s one thing that bothers me: There’s a few times where there are two ways to do something like a one line closure can omit the curly brackets, but multi line closures cannot. Rustfmt prefers to remove those brackets when it can, but I prefer to keep them, which makes editing the code faster since I don’t have a syntax error if I suddenly need a second line.

I can still live with it. And I like the clean, minimal version when I don’t have to edit. Just adding that “style” can have impact beyond how it looks involving ease of editing. And it stinks when your preferences clash with the community.

Thanks! Have you found any types of meditation work better or are easier? Suggestion where to start if it hasn’t stuck in the past?

Next time Please state that upfront: you don’t know anyone with ADHD that this has been effective for and you dont have ADHD yourself.

“Have you tried” comments from people without firsthand experience is basically asking others to be your guinea pig. I didn’t say “hard” I said “pain” as in a physical reaction. If you don’t have the same pathology I don’t think you can truly empathize or understand what you’re asking me to do (even if you think you do).

I’m glad you found meditation and that it’s helpful for you. I’m asking that you modify your evangelism in the future to give people maximum context on where your suggestions are coming from (and associated limitations).

I’m not clear. Are you someone who has ADHD or is this recommendation based on experiences with someone with ADHD?

I have ADHD and meditation, for me, is very difficult. Painful even.

Nothing about your these words reads as constructive to me:

> This is absolutely infuriating for me to see: People keep posting shit like "do it right or i will kill a puppy, you can get bitcoin if you are right" then never any testing where they change one word here or there and compare what does and doesn't work vs the dumb shit they are saying

It doesn't seem like your goal was to convey feedback in a way that was digestable, nor was in a way that was satisfiable (my previous critique).

The interesting thing here is that it breaks assumptions that to be "alpha" you MUST dominate to win. Even if within only these parameters, it suggest there are conditions where being "nice" isn't just a nebulous ethical thing, but it's an optimal conflict strategy.

I think it carries two different messages to two different groups. If you're a "lets all be friends" type, then it's important that you also guard the resources that allow you to be nice. Being provokable isn't "being mean" its the thing allowing you to be nice. If you're a "take advantage of the rubes" type, it's a hint that there might be metaphorical money left on the table by being too greedy.

> not meant to explain all and everything.

That it's not true ALL the time, is less interesting than the fact that it's true some of the time. At least to me.

The somewhat counterintuitive rules for the best expected strategy to repeated conflicts:

- Nice

- Friendly

- Retaliatory/provokable

- Clear

https://youtu.be/mScpHTIi-kM At 15:00 in.

I don’t think you realize that you’re making demands on an author who doesn’t owe you anything. And what you’re asking for is quite difficult. I think it’s okay to frame it as a wish or a desire. “I wish more articles include …”

Or as frustration on a community who keeps upvoting things you consider “not a decent article.”

Or as an opportunity “Has anyone investigated a scientific framework for modifying these prompts…”

> a decent article would have included all of that.

My retort that you can write such an article suggests that it’s more difficult than you might realize and the absence of such articles in our feeds suggests what you’re asking for might be impossible (providing hard, empirical data about a soft, non deterministic system). If you have an article that acts as a bright spot for how to write such an article, that would also be a helpful comment.

The article is stating what inputs they used and the output they observed. They stated they saw more tokens used and more time spent before returning an answer. That seems like a data point you can test. Which is maybe not the zoom level or exact content you’re looking for, but I don’t feel your criticism sticks here.

> testing where they change one word here or there and compare

You can be that person. You can write that post. Nothing is stopping you.

The problem I have with this approach is that it seems models get confused with prescriptive information such as print “it worked” or a comment stating the desired intent of code (which it doesn’t do at all). As opposed to descriptive information: generating priors and comparing them to actual output to yield either confirmation/refutation or surprise.

I think double checking is better than not, but without the ability to really “know” reason it feels a bit like adding one more hull layer to the titanic in an effort to make it unsinkable.

> there no details at all related to what changed the type of prompts etc

He gave you the exact text he added to his agents file. What else are you looking for?

If you’re a podcast person I recommend Searl’s “Breaking Change” it’s kinda him shooting the shit with himself about whatever, but I find it entertaining and informative.

To the topic at hand: I’ve not tried it yet, but I wish I could get my agent to frame everything in terms of the scientific method: state a hypothesis, brainstorm a plausible way to test it, run the test to validate itself. This is more or less TDD that I try to do with the agent, but it has no concept of that so I spent a lot of time saying “actually you can make that code change and see if it worked or not instead of just guessing that it solves the problem” it’s like an over eager knowitall junior that moves way too fast way to confidently. I need it to slow down and prove it’s not bullshitting me.

> The national-vs-local-grid topic is a red herring

I used that in passing as a measure to show how violently against regulation texas is. It was a throwaway sentence that apparently missed its mark.

> It doesn’t change anything about what I said

What did you say? I heard "we shouldn't try to make it better if it's not GUARANTEED to make it better." I countered with "it's worth trying." I think that's not your recollection of events. Maybe you're saying that the grid not being connected to the national grid didn't cause it to go down?

I'm saying that the craptastic market-focused enron designed grid system is awful and the lackluster political response afterward is not confidence inspiring. We can and should do better.

> My family and I too lived through the winter storm, going multiple days without power

That sucks. I genuinely hope you don't have to go through another one like it.

> facts

I gave you my source of facts (podcast), besides living through it. I'm not hearing different facts or sources from you. Is there something misleading or wrong about the podcast you want to highlight? Other than you're against connecting to the national grid, what are you advocating for?

> Sure but if a luxury car is safer than any other car it’s not because it was regulated to be.

"Sure but" but what? I see zero people commenting on an article titled "Safety Today is a Luxury" arguing that the only reason luxury cars are safer is due to regulation.

>So having a market where rich people can buy greater safety in the end gives greater safety to everyone.

I think this is a great pipeline, and and it's worked really well when it's been coupled with a followup: strong consumer protection regulations that make sure the best features are in all the cars.

It’s good that the industry does the right thing sometimes. It’s also good to have strong consumer protections when they don’t do the right thing.

The bill that required backup cameras came out in 2008 but didn’t fully go into effect until 2018 in order to give carmakers time. (At least) Some of what you’ve attributed to “becoming a standard” is due in part to automakers anticipating the regulation and another part of it being social norms as more and more cars have them so consumers ask for them.

I bought a car circa 2009-2010 as did a bunch of my friends (graduated college in 2008). We bought some of the most popular cars of the day, and zero of them had backup cameras.

I appreciate what you’re saying, that companies do things due to norms and markets (demand). I’m suggesting they also do things due to laws and architecture (Lessig, Code 2.0) and that we cannot rely on only one of the four to protect ourselves.

Not all auto makers just up and decided backup camera are now a standard feature. Rather: it’s a regulation put in place by Congress. To assume that cars will continue getting the same level of steady safety updates also assumes that the regulatory environment of the next 20 years looks like the last 20 years. As opposed to a regulatory environment that looks like the last 20 months.

> What I don't really understand is what videos like this tell us.

A lot of people here might intuitively understand “does not have lidar” means “can be deceived with a visual illusion.” The value of a video like that is to paint a picture for people who don’t intuitively understand it. And for everyone, there’s an emotional reaction seeing it plow through a giant wall that resonates in ways an intellectual understanding might not.

Great communication speaks to both our “fast” and “slow” brains. His video did a great job IMHO.

I was personally without power for 72 hours in sub zero temps. Every night I went to bed and wondered if my kids would be alive when/if I woke up. You know what that feels like?

I can’t do anything to *guarantee* you’ll never experience it, but I can take steps to decrease the chances or decrease the severity/dueation. I think my kids are worth it. Even if it’s not a *guarantee*.

Now that’s out of the way. I recommend you listen to the podcast. Really. Even if you lived through it. Even if you think you know everything about it. You will learn something I guarantee. It’s well produced and an easy listen. It’s an eye opener too. “The Disconnect”

Ive never taken one. They seem nice though.

On the other hand, the Tesla “robotaxi” scares the crap out of me. No lidar and seems to drive more aggressively. The Mark Rober YouTube of a Tesla plowing into a road-runner style fake tunnel is equal parts hilarious and nightmare fuel when you realize that’s what’s next to your kid biking down the street.

A few months ago I saw one driverless car maybe every three days. Now I see roughly 3-5 every day.

I get that it’s taken a long time and a lot of hype that hasn’t panned out. But once the tech works and it’s just about juicing the scale then things shift rapidly.

Even if you think “oh that’s the next generation’s problem” if there is a chance you’re wrong, or if you want to be kind to the next generation: now is the time to start thinking and planning for those problems.

I think the most sensible answer would be something like UBI. But I also think the most sensible answer for climate change is a carbon tax. Just because something is sensible doesn’t meant it’s politically viable.

I guess you’re saying that the current status is mandated by the design of the grid. Which is true, but that status would be best described as “deregulated” rather than “purposeful regulation.”

Lack of regulation and oversite around weatherization and redundancy is the main source of our problems. The Texas’ grid is market based and so unregulated that it’s not connected to the national grid so it can avoid federal regulation.

I recommend this podcast to anyone interested https://kutkutx.studio/category/the-disconnect-power-politic.... I learned that our current Texas grid was designed by Enron.

Here’s an earlier one https://fabiokung.com/2014/03/13/memory-inside-linux-contain...

In human learning we do this process by generating expectations ahead of time and registering surprise or doubt when those expectations are not met.

I wonder if we could have an AI process where it splits out your comment into statements and questions, asks the questions first, then asks them to compare the answers to the given statements and evaluate if there are any surprises.

Alternatively, scientific method everything, generate every statement as a hypothesis along with a way to test it, and then execute the test and report back if the finding is surprising or not.

Not OP but: The very essence of the program was publicity. As soon as the public lost interest (mission accomplished!), the program was canceled. We built Saturn V rockets that we never launched.

That being said, I know a lot of things were unknown. We didn’t know if the surface of the moon would interact with the atmosphere inside the capsules to combust. Some of these unknowns were downplayed. Some others were played up for dramatic effect.

The whole process of getting a person to the moon took hundreds of thousands of involved workers and the. coordinated effort of a country’s politicians and populace to fund it. I think it’s unfair to boil it down to “just publicity” but it is a big part in keeping it afloat.

As a FYI: This is unreadable by me and anyone without an X account. I see the first post and that’s it.

The input to output ratio is interesting. We are usually optimizing for volume of output, but now it’s inverted. I actually don’t want maximum output, I want the work split up into concrete, verifiable steps and that’s difficult to achieve consistently.

Ive taken to co-writing a plan with requirements with cursor and it works really well at first. But as it makes mistakes and we use those mistakes to refine the document eventually we are ready to “go” and suddenly it’s generating a large volume of code that directly contradicts something in the plan. Small annoyances like its inability to add an empty line after markdown headings have to be explicitly re added and re-reminded.

I almost wish I had more control over how it was iterating. Especially when it comes to quality and consistency.

When I/we can write a test and it can grind on that is when AI is at its best. It’s a closed problem. I need the tools to help me, help it, turn the open problem I’m trying to solve into a set of discrete closed problems.