My guess is you’re making the same mistake so many companies do. Hyper focused on tech X rather than Y or Z. Refusal to train for a month, but would rather wait a year and chase HB1s to nowhere. i.e. Looking for a unicorn instead of developing them.
I’m happily employed on the east coast working remotely now. Had to pick up a new skill or two. Never used Solr before, now I have. Took a few weeks to get proficient in between other responsibilities.
I’m glad you found a better opportunity than us.
Just maybe a note. If you read my original post, all are our existing employees or transfers.
The reality is, any good programmer can screen another programmer within 15 minutes. A conversation about there last implementation, the challanges and how they responded to them is a good start.
Coding professionally is an expression of ones self. A missing semi-column is tell, not a linting issue.
> The competition for talent is quite brutal and we just can’t pay the same as FAANG.
You don't have to pay immigrants less in order to be saving money by employing immigrants. The whole point is that by expanding the labor pool, you're reducing salaries and negotiating power for everyone.
That's not what the H1-B system was designed for. It was designed to facilitate hiring folks with specialized skills you can't get locally. E.g. you want to hire an Indian person who did his PhD on something relevant to a project you're working on.
Rather, you're leveling it by expanding the labor pool to include the rest of the world, aka via globalization. Not inherently bad imo, but I could understand why someone patriotic would find this problematic.
The alternative to fewer H1B visas is not just hiring Americans. In fact, hiring Americans is far less likely than simply hiring Indians and letting them work in India. Or, as is happening increasingly, having Chinese and Indian employees who were educated in the US, move to Canada.
Semantics. In economic terms, it's expanding the labour pool.
You could also just directly apply to a position that happens to be in Canada, e.g. https://careers.google.com/locations/toronto/?hl=en
By the way, you actually might not need any FAANG (or other Canadian employer) help: https://www.canada.ca/en/immigration-refugees-citizenship/se...
Most businesses are at equilibrium. There is no choice of "hire locally and pay them well or hire globally and pocket the difference". The choice is "hire globally and expand" (and eventually hire more locals), or "hire locally and stay small, or even get out of business".
I don't think offshoring to India is a real worry. They lack the rule of law and stable business environment of countries like the US and Canada. Nobody with sensitive IP would outsource their core business there. If offshoring worked well, even with H-1Bs being allowed, companies would already prefer keeping Indian employees there with lower salary and COL, but they don't.
I also have a hard time believing fully remote work will ever be more than a niche thing. Shared context and serendipitous discussion of ideas are so important.
If anything it will greatly help the local economy, and increases tax revenue for local, province and federal govts.
>If offshoring worked well, even with H-1Bs being allowed, companies would already prefer keeping Indian employees there with lower salary and COL, but they don't
What makes you think they don't?
https://en.wikipedia.org/wiki/Microsoft_India
Headcount is 6500.
It can just be expanded to include more jobs if it keeps getting harder and more expensive to hire in the US.
Brain drain from India to US currently hinders it, but if the rules keep getting tougher, retaining talent in India or attracting them to Canada becomes easier.
Remember when lots of people thought high quality cars couldn't be made in Japan or South Korea?
Regarding Canada becoming a hub for offshore workers, back in 2007:
https://www.infoworld.com/article/2663608/microsoft-vancouve...
A couple of years ago:
https://www.seattletimes.com/business/microsoft/trumps-immig...
2018: https://www.cnbc.com/2018/07/12/microsoft-might-be-forced-to...
>I also have a hard time believing fully remote work will ever be more than a niche thing. Shared context and serendipitous discussion of ideas are so important.
Isn't GitLab 100% remote?
I think Github has like 60% of people remote as well.
Also, the Linux kernel has been developed with people being fully remote for the most part.
These folks are getting market-rate positions -- since I'm in flyover country, there are no FAANG competitors here.
Agreed, but that won't stop companies from trying anyway. From my experience, a lot of large companies can't really tell that their remote setups aren't really working.
[Edit] Added to word 'Agreed'.
There are numerous start ups that have setup shop here (humanapi, scalable press, grainchain, brightcove, wizeline among many others).
Given the time zone closeness and culture similarities, it is normally a win-win situation.
My company even moved someone from India who lost his H1B to work here. Apparently it's easier.
> we don’t employ immigrants to save money... the competition for talent is quite brutal and we just can’t pay the same as FAANG.
No, no, we're not doing it to save money...we just can't pay the local market rate so we import cheaper workers. But it's not to save money.
Almost nobody can compete with the comp provided by FAANG. Do you want 5 companies running everything in your life? Isn’t far enough already?
Finance pays better than anyone else, and yet people apply for jobs with non-finance firms.
There are plenty of skilled people out there -- they're just getting top dollar for their skills.
The fact that salaries are so high for skilled engineers is an indication of the labor shortage.
This suggests there is no nationwide labor shortage, just a shortage of people willing to move to major tech hubs.
By allowing tech companies to evade the underlying problem by hiring immigrant labor, we are making the knock on effects (eg SF housing prices, salaries outside of SV) even worse.
Yes. FAANG engineers are good but not extraordinary.
> How many of these low-paid junior engineers are there? Enough to replace the >50% of SV engineers who are foreign-born?
Probably, if we somehow reorganized the way hiring is done, such that we were able to pick up talent anywhere in the country. There are well more than a million software developers in the US according to the BLS.
That's not available idle capacity, that's a count of people currently employed in the field including all the foreign-born developers your nativist fantasy involves replacing. And the expected 10-year grown is 24% in that number, much faster than average for all employment areas.
There are not many people that could work for us at the scale we run. [1]
[0] https://softwareengineeringdaily.com/2018/06/22/video-search...
That is over the top and provocative to the point of trolling. Please don't be this sort of internet warrior on HN. People have a right to discuss their work here without being harangued.
First you told this user that their business is unethical and should be closed, and then downthread you told them "scale is really not that hard", i.e. in addition to having superior morals you know the mechanics of their work better than they do. That is quite a patronizing double-whammy. It doesn't help to repeat "I don't mean to demean you" in comments that are in fact demeaning.
Comments like these disincentivize anyone from sharing details of what they do on HN, because who wants to be treated like that? If people stop sharing, that will make HN quite a bit worse. Therefore please don't post like this to HN.
I don't think I have words. Ethical thing would be to close? Not hire immigrants, but close. Fantastic.
> There are literally millions of skilled programmers in the US, and most of them have no interest in moving to California or Colorado.
Are they? You make bombastic statements but the reality is, that there are not that many people anywhere. There is reason why EVERYONE is having hard time hiring.
> a team of the top 50% of comp-sci graduates who can actually program couldn't solve
Programing != building for scale. This is only something experience can teach, but was majority of developers never touched any meaningful scale.
The US Bureau of Labor Statistics tracks this. I did not make those numbers up. We can certainly agree that not everyone with a degree or experience is a good hire, but good talent is not one in a hundred.
"Everyone" is not having a hard time hiring. There is a relatively decent labor market, but employers are having trouble for a variety of reasons that mostly has nothing to do with a lack of sufficient talent. Some factors: they have unreasonable requirements , such as years of experience in specific technologies; they can't or won't pay a competitive salary; they locate in high CoL cities where paying a competitive salary is even harder; they filter out people with bad hiring and interviewing processes; they demand excessive dedication (50+ hour weeks, for example); they don't post jobs where enough people are looking.
[EDIT: Another good one that came up mind: the hiring process takes too long. One of the companies I work with now takes an average of six weeks simply to respond to a candidate!! I have had several people I know through my network put in and they all got jobs elsewhere before they were even contacted! This same company cannot understand why they have such trouble finding people - they also pay well below market rates.]
I have been on the interviewer side of the table and made hiring decisions. I can tell you where I live, we don't really have problems. At my previous company, we didn't even pay well and got tons of great talent! (This did mean we had some high turnover, but there were always more.)
> Programing != building for scale. This is only something experience can teach, but was majority of developers never touched any meaningful scale.
I don't mean to demean what you do and again, you obviously know more about your product than I do, but no, scale is really not that hard, can be taught pretty readily, and the hardest problems of scaling are already solved by your cloud platform, which is half if not more of their value proposition in the first place. There are a lot of people operating at massive scales even without cloud services. It isn't that unusual.
I sincerely think you should take a very close look about what your jobs really require and what most software engineers are actually capable of doing. I've no doubt it requires skill and experience. I just don't think it requires as much as you think it does.
It also not 1 of 2. And that's quite the gist of it. We get plenty of interest, but people fall off quickly through the hiring process. We don't care what their prior experiences are. We don't demand 50+ hour work week (in fact, we actively discourage people to work past that including commute). Still takes time to find people that can learn to build for scale.
> I don't mean to demean what you do and again, you obviously know more about your product than I do, but no, scale is really not that hard, can be taught pretty readily, and the hardest problems of scaling are already solved by your cloud platform, which is half if not more of their value proposition in the first place. There are a lot of people operating at massive scales even without cloud services. It isn't that unusual.
They do take a huge load off the plate. That doesn't mean you can easily build for scale. There are plenty of companies that can put a few millions of rows in table and serve hundreds of concurrent users. Very few can do that in orders of magnitude higher. That's where we operate.
> I sincerely think you should take a very close look about what your jobs really require and what most software engineers are actually capable of doing. I've no doubt it requires skill and experience. I just don't think it requires as much as you think it does
This is hard to argue. I'm doing this for 5.5 years now, so would guess that I know what we need at this point.
This is actually not at all unusual, and even when it's not already the case, most well-architected applications serving hundreds of concurrent users can already scale to 1ks or 10ks without changing anything.
I think this is where your disconnect is. In 2019, 10ks of concurrent users is not special. It's ordinary. Even 100ks or 1Ms are not that unusual.
I am not a fan of cloud services or most of the new hotnesses in vogue today, but this is something that they've enabled even most new grads with an iota of training to do.
Looking from the outside in, the problem (and arms race) involved in determining whether one video is actually a 'like'/edited version of another infringing video seems much more interesting.
Really? So which services have 1M concurrent users, outside of the top 10?
At 100M daily users a service would get 1.1k users per sec (on average). And those have quite large deployment and plenty of engineers.
If we just think about TPS, bajillions of applications you've never heard of are dealing with large loads (10-100k TPS) in every economic sector you can think of.
Almost no games ever reach scale close to what you just wrote. The few that do are known with large technical teams.
Same goes to everything else. The reason why you can name them almost all is because it's so rare.
You are grossly overblowing the numbers. Only a very few services ever get to these numbers and once they do, they talk about the challenges of scale constantly [0] [1] [2]
[0] https://blog.discordapp.com/using-rust-to-scale-elixir-for-1...
[1] https://blog.roblox.com/2012/04/supporting-millions-of-playe...
[2] http://highscalability.com/blog/2014/2/26/the-whatsapp-archi...
Who does it benefit if his company does not exist? How does it benefit the US economy if he either hires a remote developer from another country - where all the money stays overseas - or if, hypothetically, he moved his entire company to Canada?
But your argument was that H-1B favors small companies and hurts large ones. So if Google agreed with your argument they would oppose H-1Bs.
edit: quoted doh's comment
Employers know it's harder for them to switch jobs so they have leverage. They can't complain or do much if management requires they be on-call in the evening or work late. This makes the H-1B worker more attractive to management at the same salary level, since they'll do more work and won't rock the boat. I've also seen H-1Bs receive the same base salary offer, but over time their variable compensation/RSUs (large portion of TC at many tech jobs) and salary progression are much worse.
I don't think it's unreasonable to limit H-1Bs to exceptional talent, or at least review their total compensation more thoroughly to ensure they aren't putting excessive downward pressure on the wages of ordinary middle-class citizens.
We pay the best we can. It’s completely open and transparent to all our employees. But there is no way to compete with $400k salaries paid to selected workers by few companies.
So what should we do? Close?
Sorry if this is blunt:
Paying low 100s + de minimis equity for distributed systems and DSP engineers in a super high COL city like LA is going to be a hard sell for most. People with real experience in those fields will be older, can't reasonably support a family there on that salary, and have a lot of options. The type of inexperienced employee who could pick these things up quickly is already being courted by FANGs on campus, for more money.
To get someone to work for that salary, the equity needs to be meaningful, or the startup needs to be the next Facebook or Google, not a niche product. If I'm thinking of working for you, at the max equity you list, the rosiest picture I could paint myself is an exit 4 years from now for $500mm, after which I'd get $500k assuming no dilution = $125k a year. And that's assuming all the stars align to make that happen.
I don't think you should close, since you have a cool product, but maybe consider moving to a lower COL area? I'm sure you could find plenty of people in a place like Huntsville, AL. Lots of defense contractors have engineers with the skills you need and I bet working for a fun startup would be more exciting than what they're doing now.
In person conferences can happen in LA or Vancouver with tourist visas.
And have the same time zone.
More and more companies are doing this, that's why the article mentions Toronto.
If another city say Bangalore or Vancouver does get the critical talent required to kickstart the Tech boom and be a viable competitor, tech companies will migrate wholesale and never look back.
Then this little immigration and globalization issue will finally disappear, which would be fantastic for everyone.
(2) Or we can continue opening up the globalization box piece by piece because each change is really great for some groups and really bad for other groups, which only serves to heighten social conflict and wars like the trade war that has been happening. There will never be enough assistance provided for groups that are negatively impacted by globalization; governments are much too slow acting reactively and proactively.
You realize that most of the changes that you and others want to make are just as unrealistic solutions as the above, and only one is a permanent solution? Right? And as a result of the fragmentation of the world we find ourselves in, incremental changes will not solve anything really. You can move the tech hub or dominant economy somewhere else and it will end up getting restricted again because there will never be enough relief from crowding unless the tech hub becomes more decentralized like most other industries. Further, even being decentralized there will be incumbents created in Canada that will eventually find the changes to be undesirable just like the USA.
Canadians will eventually say China is ok but Indians are externalizing too many degree education costs. And there might be another trade war, and someone thinks they have the answer by moving the dominant economy somewhere else and the same issues will surface again.. and round and round we go in circles until people have finally had enough of kicking the can down the road, throwing the garbage over the wall, and the globalization wars and option 1 happens.
>At the same time, if the companies export their labor to exploitative countries, their goods and services can be tariffed.
This isn't manufacturing. I really doubt local techies in Canada or India are being exploited.
There are many universities in Alabama and Mississippi that produce ample high-quality graduates. I can tell you that, comparing with the courses published on OCW, my curriculum and its rigor did not differ much from that of MIT.
This is where the tech companies located throughout the region get most of their employees. There are, right now, tech jobs in places like New Orleans, Mobile, Huntsville, Birmingham, Hattiesburg, Jackson, and at least a half-dozen other cities in those two states - and not just one or two employers in each, but enough to jump around a little during your career.
So, a company opening there could expect to draw from the same pool of talent everyone else is, and successfully attract a lot of candidates for salaries great for the area and substantially less offered by FAANG. Not a lot do it (just one guy in my class, for example), but as the number of jobs and their pay increased, less people would leave to work in CA. Eventually, yes, you'd drawn in people who want to move here from other places. About a quarter of my current workplace (~150 employees) relocated from elsewhere in the country.
If you get to the point where you can't find anyone else for any sane amount of money, then...open up another office somewhere else in the country. Why is your first jump that fickle immigration rules are going to be the problem rather than hiring the people already here?
Many people in smaller metro areas that you mentioned are able to relocate to the West Coast, and many people actually do. The are no visa or immigration issues for them.
If the choice for a new location is between Vancouver and Mississippi, literally anyone in the world can work in Vancouver easily, expanding their potential hiring talent pool. Whereas a Mississippi office only attracts people that are unwilling to relocate to the bigger metro areas within the US. Any new location in the US has to deal with these new immigration issues if they don't find an exact match or if they have to hire a foreign born grad from one of the schools you mentioned. Vancouver does not have these issues.
We already have sister offices in CO, MN and Czech Republic for this reason. It ads crazy complexity to the operations, but at least we have access to more talent than we had before.
[1] it's not FAANG level, but it's also not 105k. (if your top end compensation is significantly higher, you should probably be indicating that. Or perhaps breaking out minimum salary by position if they differ significantly.)
Which skills?
> it's not FAANG level, but it's also not 105k.
105k is minimum wage for an engineer. That's why it says minimum in the post.
C/C++, DBA/Architect, Java, DevOps, QA
> 105k is minimum wage for an engineer. That's why it says minimum in the post.
Perhaps it's just me, but when I see a salary listing without I high-end, I assume that's because it's pretty low.
Those are not skills, those are positions. We don't list skills, as they are quite arbitrary and really don't say much about their potential.
> Perhaps it's just me, but when I see a salary listing without I high-end, I assume that's because it's pretty low.
We found that writing the upper bound, many candidate assume that the upper bound is the general compensation and get quite upset if we don't value them as high.
We found this to be working better.
But even when "paying the same", there are other benefits of having an employee that fears they can get not just fired but kicked out of the country...
If you are worried about retention, a non-compete agreement is an option considering you are deliberately investing resources into hiring them for a position they are currently unqualified for.
I signed such an agreement and have been enjoying a nice career ever since.
Problem with training is not just cost, but mostly time. If you need to grow the business but your employees are not ready to for the scale, there is not much you can do about it. A few companies died because of this, notably Digg and Friendster.
[0] https://www.sanjosebusinesslawyersblog.com/are-non-compete-a...
Interesting, never heard about that. Any place where I can read more about it?
Friendster http://highscalability.com/blog/2007/11/13/friendster-lost-l...
There are, of course, others. Twitter wasn't far from meeting the same fate. That's (partially) why Dorsey was fired.
BTW we got RFE equally for people that are from Europe and India.
PS: because it was brought up by someone in this thread, we don’t employ immigrants to save money. Check any of my “who is hiring” posts [0] and you can see, that we pay equally across the US to all employees based on their skill level. We bring immigrants because we’re having hard times finding skilled workers locally. The competition for talent is quite brutal and we just can’t pay the same as FAANG.
[0] https://news.ycombinator.com/item?id=20587535