throwaway2016a

> so the assumption is that anyone who is an agent of Google is in on the secret

I think there is a difference here between "expectation" and "assumption".

Without the ability to do a third-party audit I agree the only reasonable assumption to make is that everyone is in on the secret and when dealing with sensitive information it should always be the assumption you go with.

However, as an expectation, I expect SaaS and social network providers (and by extension most of the HN crowd) to be better.

> There is no expectation of a video stored on a Google system being private to Google employees

"No expectation" of it not being available to "some" employees maybe, but there is certainly ways to restrict access to only a need-to-know employees. Ideally no employees at all unless some sort of automated monitoring system flags it or there is an outside report.

Just like some social networks, I would "expect" only security and moderation people would have access to profiles but there are always stories of entire companies having unrestricted access.

It's unclear from the article where the access boundaries are in this case.

That sounds like a good idea but oddly enough the first startup (the one I was at 10 years) only raised a friend and family and we were able to land customers and bootstrap from that and the second one I bootstrapped out of a little cash I got from the first one so no investors to go to per say.

Thanks for the advice, though!

This is kind of why I haven't replied directly to this one yet. I don't know how I feel, ethically, about wasting interviewers time. I've been on the hiring side of the table a ton in the last 13 years (just not on the other side of the table) and I'm not sureI'd be trilled if I found out candidates were doing this to me.

With that said, ethics aside, it does seem like a solid way to get practice.

> That should be enough to get a deeper understanding of how to design large scale systems.

Both my startups actually hit scale so I already know system design pretty well. I was actually an early adopter for K8s and we were using Docker before it was even GA. What I don't know is how to interview for it, since a down side of working at a startup is you have very few sounding boards to catch up with terms of art. The priority is on shipping. I'll check out those sites you mentioned. Thanks!

Being a startup CTO doesn't necessarily translate well to CTO (or even VP of Engineering) at more established companies. They are different skill sets. I can probably get a startup CTO / technical cofounder role at a pre-seed or maybe even seed company but at that stage the pay for execs is usually mostly equity and I need a salary to be able to support my family (that's why I'm looking vs continuing with my current startup).

I need to make $150k minimum to support my expenses (mortgage + 2 kids). My previous startup I was making $175k but my current one I am trying to raise money on (and haven't been able to yet) and the potential investors are opening saying they won't want the founders to be making more than $100k. Which is kind of why I'm looking, even if I get a seed round I won't be able to pay myself enough to cover expenses.

> Though I don't think you'll get principal engineer interviews, maybe not even staff unless you have some really notable technical accomplishments other than just being a CTO

You may be right.

The companies I was CXO of created some pretty impressive intellectual property (deep tech; which I coded the v1.0+ for personally) and two books (major tech publisher not self-published). You're probably running code I wrote on your machine if you are on a Mac or Linux. Can't say much more without doxing my throw away account.

But I'm not sure any of those things would come across well on a resume, unfortunately.

> Finally, you're going for Engineering Manager, I think you shouldn't need to do leetcode. I think most FAANGs etc are going to ask for code reviewing interviews or the like, not have you actually doing algorithmic puzzles.

Interesting, I've heard that opposite. Good feedback.

> You do want to practice system design interviews.

Good idea. What do you think the best way to go about that is?

> I think you need to figure out what kind of job you're actually looking for.

I don't want to be an individual contributor but my issues are two fold:

1. I found most management jobs want me to have managed more than 15 people and a low seven-figures budget and 15 is a non-starter.

2. I've been told (on HN, actually) a lot of companies these days ask coding questions even for management roles.

I could be wrong on one or both or those things but they are my working assumptions (the first one being based on job listings).

So I've kind of resigned myself to the idea that the only senior management or executive job at an established company.

> If you're angling for another startup gig I'm not sure why you're concerned.

I'm pretty sure (ok... 100%) I could get a leadership role at an early stage startup but the reason I'm looking is because I need a stable paycheck and benefits (for various life reasons) and most of the early stage startups I see are only paying equity with a token salary for leadership roles. If startups could pay well I'd be continuing to work on my current startup.

Sorry to hear that. Sounds like you and I are kind of in the same boat.

Unfortunately my previous employer that wasn't a startup no longer exists (it was 14 years ago after all, they got crushed by the 1..2 punch of the dot com crash and the 2008 recession).

How'd you end up pulling through it?

> Speaking from experience, I won't sugar coat it, you're not hire-able.

No sugar coating needed. That's why I'm asking this here, because I already had a hunch that was the case and figured if any group of people could have some experience with this situation it is the HN crowd.

I think you may be discounting GTP.

I've seen it notice logic errors in proprietary non-standard code that a human missed. It may not be able to literally "reason" through your code but it can follow the logic pretty well. I've even been able to have it pretty accurately comment the "intent" (vs. function) or spaghetti code with some reasonable accuracy.

This is cool but I found the UX confusing at first. It was not intuitive to me at all I had to scroll to see more information.

Anecdotally, in my architecture class (granted a decade and a half ago) we had to implement a 4-bit microcontroller from scratch. Using an actual bread board, wires, and primitive chips. We implemented those instructions from scratch given binary input switches and a manual clock. It was a simple instruction set (just ADD, SUB, STORE, RETRIEVE, SHIFT, that sort of thing) but we had to learn everything in depth.

I feel in addition to other valid comments like IP, smaller scope, etc. the simpler instruction set may have something to do with how the newer instruction sets hold your hand a lot. I always thought the point of architecture was to have to learn how things like half-adders, shift registers, and all those work and if the instruction set does it all for you, what's the point?

I had an "Assembly" course in my undergrad that taught x86, we managed in in semester. But it was a separate class from Architecture (and had Architecture I as a pre-requisite). Granted it didn't get into the hyper-optimized / specialized instructions.

I know in theory that is how it should work but is there a study to back that up?

The thing is, in my estimation, in most places the demand is so high that whatever makes the builders the most is what will get made. Which isn't necessarily what the majority of the market demands. Builders profit more off high square footage and buyers -- desperate for just someplace to live that's not the rental market -- will snatch it up. A 4 bedroom house in my market (granted one of the hottest housing markets in the US) is usually gone just days after listing.

Or, to use an analogy, if you open a water stand in the desert and sell flavored water for $5, it doesn't matter if the buyer really wanted plain water for $1, they are still going to buy your water since it is the only water available. And why start offering plain water when the flavored one sells out instantly and makes more profit?

Yes, absolutely. I did mention zoning in my message, which definitely plays a big role. Not just property tax related but also a lot of existing property don't want high density housing in their back yard.

I hear this all the time but it doesn't make much sense, IMO. Buyers do not control the square footage of inventory and for the most part are not looking for a huge house.

"The house I just bought is 4k square feet because I demanded a huge house" is very different than "The house is 4k square feet because it was one of only 3 houses put on the market this month and it was either this or no house at all."

So, while yes, it is probably true that a lot of houses on the market are probably generally larger, I'm not sure that matters for this debate unless the buyers are actually demanding that (which in my experience, they are not).

"But your houses is bigger" is no consolation when you didn't want a big house to begin with. It's a signal what the builders are building and zoning laws allow is completely out of touch with what people actually want. Yes, demand drives up cost but the demand is there because people want housing and the only houses available are huge not because they are looking for huge houses.

Looks like it's a special opportunity for students to interview early and not excluding non-students from interviewing through the normal process. Maybe they are concerned the interviews may be overrun with professional founders and they only have 1 day?

I assume it would be something like WebAuthN where it could be tied to some sort of USB key, fingerprint, or something similar.

As a technologist the most interesting part to me is they bring up "Device based age verification" and I am very curious what the current state of that is. Some initial research shows that it is only at the exploratory stage (according to my 10 whole seconds of searching) but it definitely seems interesting.

Maybe there would be an incentive for them to engage if they were getting paid for it... granted there would also need to be an incentive for them to not abuse it. But that's already an issue, for example: I had a dentist who would routinely recommend procedures I didn't need.

I know a few primary care doctors and they are all overworked and overbooked. I can't imagine any of them going above an beyond unfortunately (not because they don't want to but because they don't have the time). But most of the one's I know hate the system too.

I appreciate the clarification on tone. As a dev I read TFA in the same time as RTFM (as in: why didn't you read the article you moron?). But I get your usage as well.

Regardless of if you feel they are worthless, IMO, the people who create them should not be asked to work for free. I've never found the notes from mine to be "worthless", you may want to find a new doctor.

Also, your TFA is kind of unnecessarily vulgar (You do know what the F in TFA is for?) here seeing as it doesn't change the point of my original comment. I did, however, add the word "all" incase someone took my comment to read that I didn't think any insurance covered it.

I think the ridiculous thing here is not the doctor charging but that all insurance doesn't cover it.

I bet there are a ton of consultants and lawyers (and maybe some accountants) on HN and most of them will charge you for time spent replying to emails.

Edit: inserted the word "all" before insurance to clarify my intent.

It's purely for user experience not security. And it is pretty good at that.

Many of the methods to entering these numbers are very error prone:

- Laser scanners

- Cameras

- Humans

- Magnetic readers

- etc

The check digit lets the system know the error was likely misread / mistyped without needing a database or the like and also helps prevent an accidental mistyped value from also being a valid valud (The number for Book A is unlikely to be mistyped exactly to be the number for Book B because of the check digit).

It also can help to improve scanning speed to some degree since if you have two interpretations of the signal it is possible to infer which one is most likely to be correct.

I really like it so far.

My only gripes:

- Sometimes it can be a bit rough getting it to call my functions with the correct data.

- It sometimes won't try to search my files and makes up it's own answer (even though the correct answer is in the files)

- The outage, of course.

But I've mostly worked around that through creative prompts.

What is annoying is this isn't just ChatGTP but the whole API.

I was just getting ready to deploy the an assistant based chat bot when this happened. It underscores the importance of designing systems that fail gracefully when a service is unavailable.

The OpenAI API is currently down.

Also, the uptime graph for November doesn't look so great: https://status.openai.com/uptime

[Edit] Probably a duplicate of https://news.ycombinator.com/item?id=38371169 but I linked directly to the incident page so it didn't show up as a dupe.

Maybe I'm old but, we already had this... it was called the <object> tag and it let you embed Java applets. We also had Java Web Start.

I feel like this solution in superior to both those things but it still feels like a step backwards.