I work in health tech and I genuinely believe that this is an endemic problem that can’t be solved by a startup. It is a social issue approaching a health crisis and we need to have our policy, educator, and parents addressing this. While private industry can help, the ultimate responsibility lies with our democracy.
Why can’t it be low willpower along with availability of instant cheap calorically rich unsatiating food? In the past even if you had low will power you didn’t have access to cheap 2000+ simple carb / sugar meals. Now we do.
Many cfo actually have the biz dev unit under them. Google and my current company both have the biz dev unit under the CFO. I also worked for CFO at a mobile telecom and he was actually a key executive in charge of the 3g expansion. The role is actually much more than accounting.
I believe this little piece answers your question:
> We likely could have addressed this behavior in Image Proxy, but we had been experimenting with using more Go, and it seemed like a good place to try Go out.
At the heart of if, they were looking for opportunities to use more Go in their stack and they deemed this situation as a fit.
3. More employees knowledgeable about Go than Python
4. More enthusiasm (and therefore faster velocity) around Go development.
The blog post was about the engineering challenges they faced and how they solved them and I think it was a great write-up in that regard. The post wasn't about why they switched this service from Python to Go.
It might be, then again I see a lot of wheel reinvention in tech / NIH syndrome.
I'm the kind of hacker who if a service runs out of memory every 2 hours, writes a crontab to restart it every hour after X random minutes so they don't all restart at the same time. It gets a lot of eye rolls from the other engineers searching for perfection, but it tends to produce services quickly that are highly reliable.
And look now the engineers who like chaos monkey don't even have to set that up. It's built in.
Part of it is just Discord’s operating scale. They are already leveraging Elixir clustering to an extremely high rate of concurrency and when you start thinking about problems from that standpoint Go becomes a much more natural fit within the stack for low level micro services.
I agree that tech in general and Silicon Valley in particular has a lot of NIH, but I also think this isn't really the case here. In particular, we're discussing a Python service that performs slow image resize calls. They would have (probably, speculation on my part/experience) had to do 2 things:
1. Add profiling and telemetry to their Python code. Refactor the codebase based on insights from this.
2. Write a C<->Python interop for their image libraries.
I can't see the cost of #2 being any different than the cost they paid on writing it in Go. As for #1, depending on how the code is structured, a rewrite may have been less time than profiling spaghetti code. At that point, it depends on how much Go experience the team has.
Yeah, either a good Python JIT or Cython would have been fine honestly. I never understood the obsession with "python is slow" when you can recover almost all of the performance with a good JIT or Cython (in many/most cases).
Yes. Or simply profiling the app and optimizing sore spots would have helped too. It seems to me there was no real reason to move from Python to Go, apart from preference.
I don't think the article gives us the data to know this. Where did the latency spikes in the original implementation come from? Would fixing them have required a complete rewrite of the Python parts anyways?
I understand this is a personal preference, but having spent a good amount time with both Python and Go, FWIW I would also choose Go if I were solving the same problem.
From reading this, seems HTTP handling speed was important to them? which Go is probably better for. Also, interfacing Python to C/C++ is pretty unpleasant.
No where in the article do they mention how they factor in cost of living into the comparison. As other commenters have pointed out, most studies naively make a linear adjustment; i.e. if cost of living is 2x you need 2x the salary. This makes no sense. Say your cost of living is 30k and your salary is 100k in one city. In a 2nd City your cost of living is 60k and 180k. A rational economic decision would be to live in City 2 since it maximizes your savings. The linear adjustment based on cost of living only make sense if a person is saving 0% of their salary.
The other issue is I don’t buy the 110k vs 130k difference in Austin. They are not considering total comp in which case the delta is way bigger.
Finally, Hired is mostly used by startups and low to medium paying employers. There is a huger salary difference at large tech firms between SF and other cities.
What this study is really saying is “if you want to work at medium size companies, don’t earn anything beyond a salary, and spend all of your salary San Francisco is not worth it”.
"Say your cost of living is 30k and your salary is 100k in one city. In a 2nd City your cost of living is 60k and 180k. A rational economic decision would be to live in City 2 since it maximizes your savings."
This doesn't actually happen, except in the heads of naïve young dudes who want to justify living in SF right now.
Total REAL compensation for a non-senior engineer in SF is somewhere around $130k right now, with some Magic Stock Bux (probably worthless! maybe not!) thrown in for fun...but those don't buy burritos. Total comp of $110k sounds reasonable for Austin, as well. You can shift the ranges up or down based on seniority, but the spread is about the same, and the range caps out at around $200k in both markets for all but the most exceptional people.
In other words, the salary premium they're talking about here sounds right to me, and a premium of $20k a year is more than consumed by the 3x difference in rent, alone. You can perhaps justify going to SF on other grounds (career growth, networking, etc.), but it's not purely economically rational for most people.
Ok, your salary estimates for SF are completely off.
Let me level set by saying I went to a bootcamp and have two years experience. I make 165K (25% higher than 130) and my 3 close bootcamp colleagues all make around the same. I don’t work for some hot company like Google - 95% of people reading this haven’t heard of us. And just to be clear this is all liquid compensation (cash, bonus, and public stock equity).
So yes for me and my closest friends with comparable experience SF is by far the best option economically.
Note as others have mentioned; I don’t think that will be the case when I start a family.
"Let me level set by saying I went to a bootcamp and have two years experience. I make 165k...(cash, bonus, and public stock equity)."
I lived and worked in SF for nearly a decade until this year. I operated a company for a good portion of that time. I know what the market rates are.
Personal anecdotes doesn't mean much when the aggregate numbers disagree with you. Maybe you're special. Maybe you're lucky. You're probably underestimating what you'd make elsewhere. But I'll say this: if you're truly getting a significant fraction of your salary in liquid RSUs, you are at a public company, and are quite likely being paid above market.
If you are counting illiquid RSUs, you are doing funny math.
I don't get this thread... both cities are close enough in salary and cost that the exact economics become less of an issue. Other things, like the cities themselves and the people in them become more important.
Ok maybe I should clarify that I’m originally from Texas and work in sf so I have some pretty direct experience and know that there is a big economic difference and the salary differential is not that small.
I completely agree that there is way more to life than economics, but for the young single engineer the business case for Bay Area (and even more Seattle) is very compelling.
To put it in perspective, when I compare myself to a friend with similar skills my net worth is already 2x greater.
What? Total comp for a non-senior engineer in Austin is somewhere between $75-90k, nowhere near your $110k number, that's senior engineer territory.
Also $250-$400k is the norm for senior engineers in big companies here in the Bay Area.
I graduated from UT Austin and now live in the Bay Area, financially speaking it's the best decision I've ever made since I'm now on track to retire before 35.
Total comp for a non-senior engineer in Austin is somewhere between $75-90k, nowhere near your $110k number, that's senior engineer territory."
OK, that's fair. I haven't been a junior engineer in Austin, so I'm estimating from what I see on hiring sites.
$130k is at the top of the SF range for a junior engineer, too, so let's call it a $40k spread. That's still entirely consumed by the difference in rent and taxes.
"Also $250-$400k is the norm for senior engineers in big companies here in the Bay Area."
On this point, you are absolutely full of it. You're talking to a "senior engineer", formerly in the Bay Area. Real-dollar compensation for elite people can get that high, but most people aren't elite. Also, I don't count future value of Magic Stock Bux as salary (nor should you).
Comments about retiring at $some_young_age are treated with the same respect as when they were coming from 20-somethings in 1998: put up actual numbers or shut up. Internet People have been ambiguously bragging about their early retirement scenarios since the internet supported bragging.
A new grad offer from Google comes in at an all in compensation of approximately 150K/year. Facebook is marginally higher (or significantly higher for returning interns).
That's 105-115K base, 25-30K in stock (which is real money in this context), and 15-20K in annual bonuses. Most people also get signing bonuses that can range from 15 up to 100K (no, seriously, Facebook sign-on bonuses for returning interns are ridiculous).
All in that means 160K up to 250K in comp your first year. No, really.
Stock refreshes, laddering, and performance make it much more difficult to calculate overall compensation after that, but just as a point of comparison, the "standard" Google offer for someone graduating in 2015 (or maybe 2014, I can't remember) came with 250 shares of stock vesting over 4 years. Averaged over the past year, that's around 50K in stock. Plus a base salary that should be around over 120K, and an average bonus breaks 200K. And that's assuming no stock refreshes, and no promotions.
Granted, "working at Google" isn't average even for SF, but still.
Other than Seattle and NYC, there's a notable difference. It's not 50%, but its there. And I'll admit that Seattle is probably the best value place to live.
That said, those jobs at $bigCo also make up a smaller portion of the job market in every other city. In the bay area, something like 1% of everyone (not just SWEs) works as a SWE at Google, Facebook, or a similarly paying company. Seattle is similar, although with Microsoft and Amazon instead. While there are some jobs that pay six figures in Atlanta or Austin, they're fewer and further between.
As a result of this, I know lots of people who make 150K+ all in comp in the bay area and Seattle, and no one who breaks 6 figures in, say, Austin (all at a Junior/New Grad level).
I'll admit Seattle is where I have personal experience, and that it may not be representative of other non-SF cities.
I mean, my earlier analysis (another thread) just used the compensation figures from the fine article. Maybe those figures are an inaccurate comparisons of the two cities. If we can find a better data source, I'd be happy to re-run with those numbers. Ditto on the housing cost numbers.
It's certainly possible that one could put aside slightly more money in SF than Austin off a much bigger comp package for the same role.
Anyway, Google and Facebook are just two employers. Clearly the average junior compensation is somewhere lower. I made a lot less than that as a junior employee, but it was a few years ago, not in SF, and not for Google or Facebook.
Unfortunately there's a lack of reliable data, but from the offers I've heard, of my guess would be that someone making 100K in Austin would be making 50% more in SF, and possibly making double if they got lucky. That may in part be due to my familiarity with the area though, but again, I know more people in SF making 200K as right-out-of-college-kids than I know people making 100K in Austin.
As far as Seattle goes, I think that the state tax makes up for most of the comp differences, and housing is cheaper there, so its a better value if that's all you're optimizing for.
Which end of that range do you consider "that high?"
Anecdote for anecdote: most of the people I know who are in the 3-5 year experience range are earning closer to $200k/year than $140k. In my bracket (>10 years) it's closer to $250k than $200k for average senior developers (in which bucket I place myself).
Also $40k is not consumed by the difference in rent and taxes. At most about 85%-90% of it is (say, $20k for rent, a generous $15k for taxes).
The low end of that range starts at the high end for a mid-level engineer, and goes up into VP and director levels.
Again: I count only compensation that is liquid. Magic Stock Bux don't get marked to market, and "bonus" is a number that can fluctuate dramatically from year to year. A mid-level Googler might be making $250k in a good year, all-in, but their base salary is a fraction of that, you won't get nearly as much outside of the big tech giants.
that 75-90 seems right, but I know several in Austin/Houston that are not senior level (and have just a few years exp) making low six figure salaries. I would think the range for senior is closer to half of your 250-400, though.
However, unless you're making double in SF what you would in Austin/TX in general - I think you wouldn't be doing much better/the same/or maybe slightly worse, comparably.
If you were hired straight from college to work at a big company/one that pays graduates very competitively - then yes, SF or similar will always be better.
Yeah, it is a gold rush. It's pretty easy (relatively speaking) to make well over 200k at a public company with some experience if you include stocks. And some people are lured by options of startups (it's a gamble!). It's harder to get that elsewhere. Not everyone (or most) get there, but a lot do, and even more try.
$200k/year is a great salary, and anyone who has it should be proud. But it's not exactly "gold rush" money. You won't be "rich", nor will you be able to retire at 40.
It is entirely possibly to retire at 40 earning $200k (~ $139k net). If you spend $60k a year (I live in Sunnyvale and spend $30k a year) then you have a savings rate of a bit over 55%. A 55% savings rate has an expected time to financial independence of 14.5 years.
You could easily be a multimillionaire by 40 with those numbers. If you spent as much as me it would take about 6 years.
I don't disagree with your numbers, many people live on $30k/year by necessity. But if you do so by choice, you'll have to give up quite a bit. You won't be able to engage in much of the social activities that others enjoy in your 20s and 30s. And while it's possible (even easy) to live on $30k/yr in your early 20s, as you get towards your 30s, you'll start feeling quite a bit of social pressure from your peers to conform to a more expensive lifestyle.
So if you're willing to buck society and sacrifice some of the best years of your life, just to sit on a million dollars at the age of 40, then go ahead. But if you are willing to buck society and sacrifice so many luxuries, why do you care about money at all?
No, it is not "easy" to get that. Most Bay Area jobs are not senior engineers at Google. Like I said before, a mid-level salary in SF is somewhere around $130k. The numbers are well-documented.
If you win the stock lottery, or stay in the industry for a while and become one of the top 10% of earners, you can do upwards of $200k. But that excludes the bottom 90% of jobs, which is what most people actually have.
As a PSA to all junior engineers: if you actually believe you can "easily" earn over $200k in the bay area, I strongly encourage you to take some phone screens before acting on your notions. Tell companies your salary expectations in advance.
I agree with you and don't believe the high salaries reported here. The best reference set is the H1B salary data which trends with the glassdoor self reported data.
The folks making closest to 200k or higher are largely Computer and Information Systems Managers in Santa Clara county at levels 3 and 4. I imagine these are Senior Directors or VPs.
The argument is that the expected value of salary + rsu/stock/options is effectively just salary. These reddit posts basically agree with H1B data. Salaries are the bar. A H1B has to be paid more than the prevailing wage for the position and the company has to have documented that they could not find a qualified American for that role. Prevailing wages are set as a function of the H1B + American workers so over time salaries rise.
But that's not the case, since salaries are significantly less variable than stock comp.
For high paying jobs, salary is 50-80% of total comp. For lower paying ones, it's 95+%, which is what the salary surveys show.
So comparing salary only, you strongly bias against top paying positions since salaries make up a smaller portion of the compensation for those positions.
You should wait till they cross the intersection instead of timing your cross with the pedestrian. You need to yield to the pedestrian as soon as they enter the cross walk - that's the law and what all road traffic is required to follow.
As others have said, biking infrastructure in NYC (and in basically all U.S. cities) is an afterthought. It's been tacked onto an environment that was totally given over to cars (rather absurdly, given New York's status as the only truly pedestrian city in the U.S.).
What you're seeing is the cyclist's attempt to navigate an environment that's directly hostile to their interests. As long as nothing about that changes, this is what it's going to look like, because I'm not going to follow rules that make no sense for my mode of travel.
With respect, of course, to the fact that this is an unpopular opinion in some circles, I'm still just not going to do it. And, crucially, I'm comfortable with the social and legal consequences of my position.
I think people forget that vc backed startups are profit and greed driven (like all free enterprises). We all like to drink the cool aid and think that a startup's purpose is societal impact but really they are engineered to make $$$.
As such, it makes sense that people pour millions of dollars to target the 1%. They've got a shitload of more money than the rest of the country. Seriously, if you've got an product that every rich person wants but that no other person cares about, you will have vcs knocking on your door. We need to think of vcs as building cash machines, not as the money behind "changing the world". Yes occasionally vcs fund companies like Google that have broader impact but that's not their primary aim.
If we want more people building ventures that impact society we need to either
* change the funding model
* build bootstrap ventures
* build non profit ventures
* ask the government to step in (a socialist approach)
A good example are institutions like the Gates foundation and WHO eradicating polio in the last few years. Yeah it's not sexy but organizations like that are making real societal impact.
Facebook, Instagram [and any similar VC backed company] are not targeting the 1%. Quite the opposite. Mass human scale adoption is their goal. And once this is achieved then their biz models get very interesting. I do hope there is a significant shift towards social impact ventures. This is where some of the hardest challenges are.
That sentiment is commonplace enough that Silicon Valley parodied it: "We're making the world a better place. Through constructing elegant hierarchies for maximum code reuse and extensibility"
Something that's been on my mind recently, given the increasing amounts of inequality, is "why aren't the lower and middle classes more effective at taking rich people's money?".
In other words, rather that thinking about protectionist measures or redistribution (not that I'm particularly opposed to such things), why aren't rich people spending more cash on things and thereby allowing the money to flow down?
One possibility is that we really are topping out our hierarchy of needs. Billionaires are driving round in Priuses and wearing $80 jeans. Capitalism relies on demand being practically infinite. Could this growing inequality be a sign that this assumption is faulty?
Or do we just need to invent more drone mounted, blockchain integrated juicers to sell to them?
The reason that lower and middle class people are not more effective at taking rich people's money is because if they were effective at taking money, they would not be lower and middle class. They are lower and middle class because they are effective at other things that have nothing to do with taking money.
Perhaps I should have phrased the question as "why have the lower and middle classes become relatively less effective at taking rich people's money in recent years".
This is not how most large companies I've worked with operate. Large IT decisions are driven top down, not organically through the engineering front line. The CTO, CIO, etc. work with vendors that market to them. It's the way organizations like Salesforce, Workday, Cisco, and Concur operate. The consumer/developer first organizations like Dropbox, Twilio, and Stripe can only get to a certain stage before they need to target the big fish. It's this huge cultural shift (focusing on making something users want vs figuring out how to sell to behometh organizations) that will make or break these companies.
If they figure it out, they will own their markets because frankly their products are better since they were user first.
I think it's important part of building a company - recognizing that the heart and soul that led you to say $100M sales can completely differnent from what leads you to $1B. It's almost like a mini innovators dilemma for startups.
Extremely well said. Yours is a more in depth and eloquent version of what I wrote in the first section of this essay:
"This is an essay about go-to-market strategy and market development. It’s also an essay about company culture.
Specifically, it’s about how the market focus and culture that helped a company reach significant heights can rapidly transform from critical assets into potential liabilities…and what to do about it.
While Twilio is the focus of this essay, this essay is not just about Twilio. It could be about about ANY potentially disruptive company with brilliant founders, venture-scale ambitions, great products, a top-notch team, and traction to die for."
Yes, absolutely. However, does this change over time? With:
+ increased decentralized decision making
+ less monolithic systems architectures and complex integrations
+ quicker/cheaper ability to take software from pilot to production
Do developers become the new gatekeepers of the large enterprises?
"Do developers become the new gatekeepers of the large enterprises?"
I doubt it. From what I see is that developers can do a few things but as soon as the IT department hears about it they will try to standardize on a solution. And you are back in big Enterprise sales. In the end the CIO wants to keep his/her budgets and power.
I just went through this. We prototyped a system and thought it would work. IT jumped at it and made that supplier the company wide standard. In the meantime we got some doubts and will probably jump ship. But IT is sticking to their decision.
Most examples (Stripe being the exception) are for internal use, and I think there is a difference between software used internally and API's / libraries used in the product.
Stripe is about payment and payment will be at the core of most businesses so it makes sense for the higher ups to take an interest.
My partner is a doctor who works at one the top medical systems in America. The EHR system she works is atrocious. Last time she asked me to look at it to help figure out how to print an image and I stared at this 90s era windows app that looked like the pied piper interface. I gave up after 15 minutes and we ended up screenshoting the image. Taleo felt like it had a better UX than it.
I can't believe we let some of best doctors in the country waste hours each day fighting with terrible software instead of treating patients. My partner spends more time trying to figure out the EHR than she does treating patients.
