Valuing stock grants is terrifically complex, particularly at companies that aren’t publicly traded. I don’t have the time or inclination to do the math, so I use massively simplistic heuristic:

At publicly-traded companies, I consider stock grants to be fairly reliable income, and I value them at roughly the present-day value. If I’m offered a job at Apple that includes 100 shares, I can look up the stock price ($143.50 as I write this) and figure what they’re worth today ($14,350). Stock grants usually start vesting after a year of employment, and so, sure, I can’t know what they’ll be worth next year. But short of some catastrophic event, I can be pretty sure they’ll be worth something, and probably something within, say, 25% of what they’re worth today. So I might discount stock grants at public companies a bit, but not much.

At private companies, I treat grants as essentially a lottery ticket, and value them at $0. I find private stock grants overwhelmingly complex, and incredibly far from actual money I can exchange for goods and services. They also come with some serious downsides, like needing to spend money to exercise options, and pay taxes, before you really know what they’re worth. So at private companies, unless there’s a very specific plan to go public soon, my rule is to treat options as a gamble and don’t consider them part of total compensation.

jacobian.org/...​ct/13/tech-salaries-2021

First: stock options are BORING AND COMPLICATED AND AWFUL. They are full of taxes, which we all know are awful.

jvns.ca/...​th-on-your-stock-options

cletus 12 days ago | parent | favorite | on: Top Paying Tech Companies by SWE Level

As a FAANG engineer going on 9 years now, let me address the usual rebuttals:

— There is a selection bias. Nope, this is pretty much accurate.

— These aren’t real. Yes, yes they are.

— Self-reporters are lying. Maybe some do but these numbers are pretty accurate. If anything, I question Lyft and Airbnb as such outliers. I wonder if this factored in Lyft’s post-IPO stock performance and makes unrealistic valuations of Airbnb’s RSUs/options. But for any listed company, these numbers are accurate.

— You have to work incredibly hard for this compensation. no, you don’t. In fact you’ll typically find significantly better work-life balance at a FAANG than a startup.

— These numbers are inflated by years of stock growth that is unlikely to continue in the future. This there is some truth to but not as much as people claim. Amazon, of all these companies, builds in expected stock growth into their initial grant valuation (which I think is total BS; if any Amazon recruiters are reading this, please stop). But I know what offers new hires can get pretty accurately so at current stock prices as a new hire these numbers pare pretty accurate.

— Newer offers are likely to be less. False. If anything, initial offers continue just climbing such that anyone who is interested in maximizing their compensation should probably move companies every 3-4 years, especially 4 if you don’t get an additional grant after your initial grant has fully vested.

There are some things to be aware of though and these can make it nontrivial to compare competing offers. Some examples:

— Most FAANGs have a 25/25/25/25 vesting schedule. Amazon does not. It’s vesting schedule is 5/15/40/40 with a vesting signing bonus in the first 2 years to (partially) compensate for this.

— Amazon, as noted earlier, assumes stock price growth in their offer.

— Amazon (noticing a trend?) has vesting on 401k matches that can take 2-3 years. Most FAANGs do not.

— Anything less than a 50% 401k match is below market.

— Some FAANGs have caps on 401k matches. Some don’t.

— I think the most generous 401k match I’ve seen is Google’s at 50% of your contribution with no cap or vesting period or 100% of the first $3,000 at year’s end, whichever is higher. The really nice thing is because there’s no cap you get it immediately. It’s fairly common to get your bonus in January, put it all in your 401k, get your 50% match and you’re done for the year.

— Some offer the ability to make contributions into after tax 401k (Google “mega backdoor Roth” if you’re interested in this). This is potentially huge beneficial. You can use it to invest money you can withdraw at any time at no penalty but the investment returns are tax free. If you withdraw the returns (not the initial investment) prior to being aged 59.5 you pay taxes plus a 10% penalty, however.

— Vacation days vary but 4 weeks (20 days) should be considered the norm for the US (30 for Europe/Australia).

— Some companies (eg Google) start you on less vacation days but you get more with length of service.

— Unlimited time off is bullshit. Think of this as no time off.

— Health insurance can differ but I imagine pretty much all FAANGs at this point have good health insurance. The gold standard is probably Kaiser for CA residents.

— Some FAANGs have a 1 year cliff. Some do not (eg Google, FB).

— Vesting schedules can vary. Some are monthly, some every 3 months, some annually. Try to avoid anything less frequent than once every 3 months. It can create bad incentives for the company to get rid of you before a big vest date.

— FAANGs will give you performance-based RSU grants annually. The time of year can vary. The eligibility can vary. For example, Google gives you a refresh grant at, after Q2-Q3 calibration (based on your previous two halves). And I believe in recent years it changed that if you joined that calendar year you aren’t eligible.

— Because of refresh grants and your initial grant running in tandem, years 2-4 can often be your most lucrative. If you don’t get promoted or an additional grant you can get significantly less compensation in year 5. Why these companies let people leave because they won’t give them additional equity rather than competing for a new hire is beyond me. But they do.

— Because of the inflation in initial offers, a new hire can often have a significantly better offer than someone who joined 3 years prior. The veteran may only have higher total compensation because of refresh grants and/or stock growth.

— FAANGs tightly control salary within bands for a given level. Going beyond this is unlikely to happen however there is FAR more movement on RSUs in an initial offer and/or signing bonus.

So this is all another reason of why from a financial POV working for a startup is—how should I put this?—suboptimal. Your equity is probably worth nothing (even if you get acquihired, liquidation preferences probably mean all non-founder stock is worth $0). The hours are worse. The benefits are worse. There may be reasons to do this that aren’t financial (as a non-founder) but personally I’d suggest people use their most productive years to ensure their financial independence and then chase whatever moonshot tickles your fancy without the pressure of having to pay for food.

I had a recruiter call me a year ago with what sounded like a really amazing offer. Working for a well funded startup on a problem in which I happen to be a domain expert.

They’re aren’t many expects in my field with my background and skill set and they said the right things to get me to interview (compensation won’t be a concern).

Anyway... Went in for 4-5 interviews. Took me about 10 hours in total to interview with them.

Kept hearing that they have a difficult time filling the position. Reports directly to the CTO, etc, etc.

They came in at 1/2 my current salary. It was about 30-40% of what you would pay a decent engineer with experience.

I told them my salary level BEFORE the interview...

Gave them a hard now. They came back and said they can offer more stock and remote work.

Why in the heck would I turn down a bird in the hand for 1x in the bush?

Part of the problem is that if you’re interviewing you need to know what you’re buying!

You can’t go to a ferrari dealership and offer to buy one or $20k.. that’s just not how it works.

I had the same thing happen to me a couple times. Long and tiring process. Blackboard, make a project, lots of dirty laundry about “long gone incompetent developers” that made the current system a giant mess. Lots of red flags but lots of promises including a large maximum salary, including bonuses and stock options.

The final offer, however, was 60% of my current pay. Lower than the average for someone with my experience and position.

I had good rapport with the HR recruiter and flat out asked why they proceeded, even though she knew I was already getting more, at a better company, with more benefits.

The answer: they assumed I was lying about my current salary and I was just bluffing. Recommended me that I lie in my next interview.

news.ycombinator.com/item?id=19707543

Secret robot internet
www.reddit.com/...​he_secret_robot_internet

stripe.com/atlas
news.ycombinator.com/item?id=11166417

По инфе из тарифов Unicredit, они перестали работать только с USD-чеками. EUR, вроде, еще берут. Возможно, еще кто-то остался. Плюс специализированые компании (не банки) разной степени стремности.

dou.ua/forums/topic/15437
www.unicredit.ua/...elers cheques_03.2016.pdf
minfin.com.ua/2016/04/20/17900583

from __future__ import braces

www.quora.com/...ode-review-policy-process

1) www.reddit.com/...ternet_fucking
2) askville.amazon.com/...questId=2956513
3) 600-1000 ms
en.wikipedia.org/...#Signal_latency
arstechnica.com/...cy-still-awful
4) www.dslreports.com/...-your-satellite
5) www.sputnik.datagroup.ua/ru/prices
6) не працює (нормально)

Original complaint is from ~Dec 2012, but it looks like the lawsuit has been lasting for a few years.

dou.ua == developers.org.ua.ua.org
Дальше douuaorg.developers?

Ещё бывает ООП головного мозга.
Как два программиста хлеб пекли

I don’t always test my code...

Ось вам пиріжок. Тільки він без варення. Та, мабуть, і без тіста. Та й не пиріжок, а бублик. Власне, дірка від бублика.

И вот сегодня пришло письмо: “вам добавили $Х00”. Все.

При этом за год в моей работе не изменилось абсолютно ничего.

Ви стали свідком покращення

3.7X speedup from removing a call to sleep

Client gives me a Help Desk Ticket. User claims batch job use to run in 5 minutes but now runs for hours. The logs confirm this. The commit logs show that an offshore programmer forgot to remove a 10 second sleep command from inside an iteration (for debugging I presume) before promoting to production. I removed it and got a 100X improvement in throughput.

My client said that now his user loves him; what did I do to fix it so fast?

When I told him that I removed the Sleep, he said, “No! No! No! Change it to a 5 second Sleep so I have something to give him the next time he complains!”