Hacker News new | past | comments | ask | show | jobs | submit login
Show HN: Visa Explorer – Salaries, lawyers, companies of all H-1B in the USA (jobsintech.io)
582 points by negrit on May 22, 2015 | hide | past | favorite | 247 comments



Be aware this shows the salary as it was when the visa was applied for; which is often many months before the visa (end supposedly, the job) begins.

I found myself on here (the only H-1B at my company), and the salary is almost $20k less than what I was being paid by the time the visa stamp actually arrived, and $80k less than today (3 years in).

As frustrating as actually going through the US immigration system is, I do appreciate that they put out some good data on it.

I'm currently going through PERM (and again, frustrations, audits, and personal nightmares aside), it's great to see that I can look at all the processed applications from just 1-2 days ago - and see details on company/role. There are plenty of sites that the ugly/difficult data that govt puts out itself, and makes it a bit easier to navigate; one example I've used is dolstats.com


Income can be skewed, but also note that employers will often under-report the level of the position that the applicant is working.

This is part of the strategy to get through application approval, since one of the criteria for approval is whether or not the person is making more than the average salary for citizens working the same SOC job in the region.

You can read more about SOC and see the raw data (disclosure data) that was used in this tool here:

http://www.foreignlaborcert.doleta.gov/performancedata.cfm


interesting. is that why google only has the main description? almost everyone else list some level along with "software engineer". google doesn't even risk saying "sr."


Is it a good indicator of what the company pays entry level positions if the company mostly hires for one specific job. So would you use the Google, Microsoft, etc data as a good indicator of the average start salary of an entry level dev today?


H-1B salaries are actually "pegged" according to the title of each position; they call this the minimum or prevailing wage, and it's specific to an area.

For example, if a company is hiring a "senior software developer" in NYC, and the US government considers $120k to be the prevailing salary of a senior software developer in NYC, then the company must pay that person at least that much; Google can't pay any less, even if it wanted to. The wage database is public [1].

So that gives you the minimum, at least.

[1] http://www.flcdatacenter.com/OesWizardStart.aspx


Technically correct, but as we all know, the title of a job doesn't necessarily correspond to the job responsibilities. Scope creep is just as real for employee positions as it is for company projects/products.


True. I don't know how much the US government scrutinizes applications (which have to include a description of the position, responsibilities, etc.) and applicants. Perhaps not at all.


Yes, my job title on my US Visa was first Computer Systems Analyst, now it's Software Engineer. My college's who is also on H1B is Data Scientist, despite doing exactly the same work as me.

Our immigration lawyer decided on this job title, I didn't even know I was going to be Software Engineer before I got my visa. It's just that my college has a PhD and I don't. So, his profile fits that of a Data Scientist.


Data scientist has a lower prevailing wage.


the last level for sofware engineer in california

Level 4 Wage:$52.80 hour - $109,824 year


It's not just that, looking through my current and past entries some of these records just aren't correct, and don't even match the data provided on LCAs etc.


Hello everyone. I quickly built this little tool based on public records provided by the government.

The amount of information they provide is incredible. There are no explicit names, but for small startup or exec position it is easy to guess who is who.

As far as I know this is the biggest LCA database so far.


Thanks for this! I'm currently looking at opportunities and this would help immensely during negotiations.

Would it be possible to search for role + location instead of company first? Essentially the refine functionality you provide on company's yearly page [0] to be at the root. You can perhaps make state or city mandatory so that it doesn't hit your backend hard.

[0]: http://data.jobsintech.io/companies/***/2015


Try http://visadoor.com

You can also search Green Card PERMs.


Yes, This is definitively something I'm going to provide.


Excellent.


Their data hadn't been updated in a while, but http://www.salar.ly/ has that functionality


Great implementation. I've seen the raw data before but never had the web development skills to execute on this idea. Can you also add a query where once you select a company and year, it groups by title and gives average salary.


For a company and a given year, you can search by title.


It would be useful to search across companies for a job title


I thought clicking "Jobs" would lead me to something grouped by job title...


Can I find the highest salaries?


Take a look at Airbnb, Data Scientist. This person has been doing >$1.000.000 a year...


I've seen that. It's either mistake or he sold the shares.


what does "status: withdrawn" mean ?


It means the petition was withdrawn. My guess is that that person left the company.


I actually think it means they decide to no make go further with the LCA. Like they decided to finally not hire the employee or the employee decided to not work for this company anymore.


Actually that makes more sense. It is possible to be working on another visa while a company files the LCA for you (TN, L, visas, etc), but it's probably less likely than a new H1-B hire.


They probably misfiled the LCA at 1m and re filed at 100k.


Netflix? Senior Software Engineer, $375K.


Serious question.

What kind of skills of knowledge or experience do you have to have so that you take more than quarter million of dollars??? Incredible!!


They ran the million dollar challenge[1] to increase viewership, thus decreasing churn.

If you could (make a case that you) consistently decrease churn then you'd pay for yourself at that rate.

1: http://www.wired.com/2012/04/netflix-prize-costs/


For a company like Netflix, my guess is, you need to be a world-class expert at some area of scalable backends and/or big data. For example, you need to be a core contributor (in terms of concepts, architecture etc. - not just coding) to projects like Kafka, Cassandra, Spark etc.


Very impressive. I'm organizing an Immigration-themed "Startup Weekend" event[1] happening next week (in SF), and I think I'll be sharing this with participants there as an example of what can be done.

[1] http://www.up.co/communities/usa/san-francisco/startup-weeke... (Use promo code "hn" for 70% off)


Definitively. I'm going to provide an API pretty soon. Too bad it wont be ready for this week end.


Neat project. My brother, who is an immigration lawyer and presumably somewhere on the list, asks for a way to search by attorney.



Instead of plotting a count per exact salary could you plot count per binned salary? Ie instead of plotting 1 at 105k and 1 at 107k plot 2 at 100-110k.

As it is the plots aren't particularly useful for comparing the distribution of salaries between companies etc.


Thanks this is way better than existing sites.

I did notice that searching for Google gives you 5 different entities that are all Google.


Yes I am still cleaning the data. It's all unbelievable the amount of crap you can find in the public records. I have a lot of duplicates unfortunately.

Down the road, I'm planning on providing a couple of cool charts/map about trends and evolutions.


A few things I'd be interested in:

- ratio of LCAs to total permanent full time employees. This will show which companies are really leaning on the H1-B visa (I'd estimate my "household name" software company is at about 50%)

- Source of prevailing wage. Interestingly employers don't have to use the BOL published data and can self report. I'd be interested to see how many self report.

- The average delta between prevailing wage and salary per employer/job title.

- A way to search by geographic region.


Excellent ideas. Hope to see these things soon.


Check out Open Refine. Has a feature that clusters similar strings and unifies. I remember last time I looked at this data set... 4 letter acronyms spelled 12 different ways, it's unbelievably messy.


For search result of a company name, can you sort the results descending by # of LCA? this would put the most likely true result of top, I think.

I also noticed that the sort buttons (for salary) behave backwards of what I'd intuitively expect?

Would it be possible to auto-merge likely duplicate company names?

Great job, keep it up!! :)


I'm currently coding both feature. :)



Any plans to open-source this on Github and accept pull requests?


Great work. Impressive little tool.


This is a great tool. I hope it helps every H1-B fight for the skills they are worth.


May I ask which public records did you use? Thanks, this looks awesome!


can you add functionality to search by job title other than as a filter? great tool though!


Yes, work in progress!


I would like to sound a caution, if you are using this data for any kind of salary negotiation.

- LCA for a candidate is filed only once per company as long as he is in the same job role. Factos like getting promoted to Senior X, recent pay-hikes would not get reflected in LCA.

- Wage listed may not include bonus, and other compensation. So dont leave money on the table while negotiating. Crosscheck with glassdoor.

- Companies also go through frequent job title changes, so it is possible you might miss more dataset hidden in other job titles.

- When you are looking for wage related idea, only latest year matters.

- Some people have mentioned earlier that H1B have 10-20% less salary. My personal experience says otherwise. There is always bound to be variation even among non H1B.

Finally - I am not a lawyer, and none of it are legal or professional advice.


> Wage listed may not include bonus, and other compensation.

Yeah, once you factor in RSUs and bonus my actual salary is about 40% higher than what was listed on my LCA.


I work in finance, and browsing this list reinforces the assumption that people on H-1Bs are just cheap labor for employers. Most people on this list whose positions I can determine with confidence seem to make 10-20% too little.


I'm a software engineer, and I searched my company and found that the average salary listed was inline with what I know to be our baseline staff salary for developers. (For what it's worth, we annually review industry salary surveys and adjust compensation for our existing employees accordingly.)

A lot of engineers (especially in the Bay Area) either want the excitement and IPO potential of a start up or the stability of working for a large, established company. It makes recruiting difficult for mid-sized companies like mine.


Given the high salaries of tech employees and the low unemployment among them, I think that "cheap labor" isn't exactly what I would call the H-1B system.

It is a bit exploitative of the immigrants, by which I meant that the companies use the strict/stupid US immigration system as leverage. But citizens or taxpayers in the US have no reason of complaining. Maybe these H-1Bs are preventing an entirely cut-throat competition for talent, but on overall, those people add to the economy, and in a big way. Anyone who earns like 100K a year creates numerous other jobs and pays a lot of taxes...


Check Netflix. Most of their H1Bs are making far greater than prevailing wage for their positions.


Yeah, is this data right? Why do they have Software and UX people making over $250k?


See my reply below. TLDR version: Netflix compensation is all salary. Google, etc, comp is salary + stock + bonus, and I suspect these applications list just salary, not total comp.


Isn't it the case that compensation at Netflix is variable and you could select how you want to distribute it over salary/stock/bonus?


Yes, that's true (for at least stock, I don't think they do bonuses). But the default is 95% salary / 5% stock, so to simplify, I said "just salary"


It's fairly common knowledge in Silicon Valley that Netflix salaries are insane.


Isn't that because they hire a very small amount of very experienced people?


Or maybe because of their high turnover, to compensate the risk of what is known as "the culture of fear" there.


For $300k+ I'll take on some risk.


If you're a US person, why not... But for someone who needs H1B sponsoring and relocation (probably with a family as well), Netflix is the worst company to work for as your first American employer.


Why, specifically? ... I came to Netflix on a H-1B, and they have been awesome. But if there are specific concerns (eg, no grace period in case of termination? something else?) I'd be interested to know what they were.


Because initially H1B's are tied to your employer. I think for the first 2 years? So if you're let go during that period, I believe you have to move back to your country. You can't directly transfer H1B to another company [during that period].


Disclaimer: I'm not an immigration lawyer... So while employers sponsor H-1Bs, and the employer name is associated with the H-1B, you shouldn't be "tied" to the employer, certainly not these days post AC21 (http://en.wikipedia.org/wiki/American_Competitiveness_in_the...). I've been headhunted while on a first term H-1B, and recruiters didn't blink when I mentioned I was on a H-1B. Transfers are common. (Has an employer given you the impression you were tied to them?)

But you mentioned being "let go" during that period, and needing to move back to your country. (A problem created by the H-1B's lack of a stipulated grace period.) Is Netflix considered a poor choice as it's seen to be quicker to fire, therefore this situation is more likely?


Personally, I was considering Netflix as a poor choice as the first H1B sponsor because I thought you (your H1B) are tied to your employer for some initial time. If that is not the case, I could reconsider Netflix :) Assuming that they give enough termination notification (and also helping with landing new jobs, as they mention in those famous slides). My former boss works at Netflix for a couple of years now and mentioned about good opportunities there, but Netflix was out of the table for me because of this possible initial H1B moment and generally because of that "the culture of fear".

Btw, how do you feel about this? Does this fear really exist?

Is it possible for a good engineer to have a relaxed life and atmosphere and a family friendly situation at Netflix?


I know from personal experience that that is not correct. There is no such restriction.


It's also because they explicitly prefer to shift as much compensation as possible towards salary as opposed to stock, benefits, perks, etc. They released a long presentation about why they do this, I think it's somewhere on Slideshare.



Netflix engineers are better than the average engineer (even in a given job title), so the prevailing wage calculations don't represent the true economics.

But in any case, H1-B visas can lower the wages of engineers without H1-B holders earning less than Americans. It's just like any import. The import itself doesn't have to be sold for cheaper, it's just that the presence of the import lowers the price via competition, so that both the import and the local good are cheaper than the price before importing started.


Doesn't seem to be the case for engineering positions, from checking out a few employers in the bay area.


Looks as if you are looking at some different data set than we all are. Check out Netflix, Google, Twitter etc. - these are more than competitive rates.

Unless you went into bodyshop companies like TCS, Infosys etc. to make yourself feel angry.


Infosys seems to classify everyone as a technology lead. They don't have any software engineers!


All imports are "cheap" whether labor, steel or iphones. Cheap is good, because if you can get something cheap it means you can get it with minimal resources.

The only reason to limit immigration is externalities. For example, Israel has very open immigration for Jews. It does not allow open immigration for African refugees because they believe it would have a negative impact on their society. For some strange reason, other nations are fearful of stating the same thing, but they clearly have something like this in mind when they limit immigration.

Cheap labor is a net positive. The only valid reason to oppose immigration (and I think it is very valid) is the negative social externalities that immigration causes.


This data doesn't include non-salary compensation (e.g., stock grants). For tech companies, that can be substantial.


There is a Senior Software Engineer at IBM recruited in 2014 who earns $7,278,870,000? [1]

How do I get this job? :)

EDIT: The application was withdrawn. That saved them some money.

[1] http://data.jobsintech.io/companies/ibm-corporation/2014?ord...


Decimal place error?


Looks like visas are made for Indian companies and not american companies http://data.jobsintech.io/companies


Darnit, I missed out on the $43,264,000 postdoc at MIT in 2008.

http://data.jobsintech.io/companies/massachusetts-institute-...


Such a salary in academia is really incredible. No way this is not just a typo?


Definitely a typo. A programmer at my company doesn't make 7,600,000.


For a post-doc? That's likely $43,000.


Wow, you aren't kidding when you say Netflix pays well!


Oh my god, $350k for a senior software engineer? I'm absolutely floored.


That's not an unreasonable total compensation level for senior engineers in the bay area.

Remember that Netflix does not typically offer bonuses, stock grants, etc, and just pays people a salary. While most other big tech companies (eg, Google) have salary as 1/2 to 3/4 of your total compensation, with the remainder being bonus and stock grants (or options). So in the Google table here, you will see far lower salaries than Netflix, just because at Netflix, your total comp is almost entirely salary. I suspect that the Google H1B's are getting stock grants and bonuses that will increase their compensation substantially.

I'm not an H1B, but I'm speaking from experience. I'm a senior software engineer at Google, and I'm starting at Netflix in a few weeks. My total comp will be in the same ballpark at Netflix as my current total comp at Google, but will move from salary + bonuses + stock at Google to being almost entirely salary. I'm excited that I might actually be able to do my own taxes in the future.


Reasonable would be anywhere close to average. $350K is very much an outlier. It has to be. For every "senior software engineer" you show me making $350K, I'll show you 20 making less than $100K. $350K is insanely great, we're talking top-2% territory.

This seems to happen all the time in the comments here. Someone points out some insane top salary they found at some insane top company, and a bunch of people chime in with comments like "Oh, that seems pretty reasonable actually". It doesn't seem reasonable. It's not anywhere close to any median or mean. It's the far right of the distribution.

Find any company on Glassdoor where $350K is anywhere close to the center of the bar for "senior software engineer". I'll then eat my words.


So Senior Software Engineers at Google make about 250k in total compensation.

http://www.glassdoor.com/Salary/Google-Senior-Software-Engin...

Now, 250k isn't 350k, but 350k is the high end of Senior Software Engineer at Netflix, and that's what Staff Software Engineers at Google make.

http://www.glassdoor.com/Salary/Google-Staff-Software-Engine...

Staff Software Engineers are "extra senior engineers" that are on the same HR level as engineering managers. I don't think Netflix makes this distinction.


Isn't it better the Google way? You will be taxed less since you can keep the stock for 1+ year and get taxed at the Long Term Capital Gains rate. Being almost all salary will get you taxed to death.


My Google stock comes as RSUs (well, they call them "GSUs"), and it is taxed as if it were income as it vests.


Options gets hit by AMT tax when you strike them(assuming your total income puts you into AMT). You're going to be screwed either with RSUs or options by taxes just little less with AMT.


not so because options and RSUs mean that until the vest you don't get ownership of the stock. So if you keep it for an ADDITIONAL year then you get taxed long term on the _additional gain_.

Basically options and RSU = normal income


Compare them to Microsoft for equivalent-sounding positions in CA, interesting...


Really not that much when you take out the taxes, food and housing premium of the area. That is what should floor you as well


You are utterly and completely incorrect.

$30,000 a month is a fantastic salary. No ifs, ands or buts.


Yes, considering that is the equivalent of a minimum wage yearly salary every two weeks.


    Monthly Gross Pay $29,166.67
    Federal Withholding $8,195.61
    Social Security $1,808.33
    Medicare $422.92
    California $2,831.70
    SDI $262.50



    Net Monthly Pay $15,645.61


5k can get you a nice place in SV (maybe even less)

Even if you have kids you can probably save something like 5k/month, which is good.


it's better than that because social security tax will be capped four months in.


It is still pretty good. If you figure the housing premium between the Bay Area & a "normal" city is $1.5M vs $300K for roughly the same "middle class" house, then the monthly mortgage difference is about $5500, or about $66K/year.

So that $350K is equivalent to $284K in a "normal" place. Even being really conservative, and figuring extra costs (private school tuition, because even that $1.5M might not be enough to get you good schools, extra CA taxes, higher insurance rates, etc), we're still at an equivalent of about $250K. And that's a lot better than most people could get in any "normal" metro area as a Software Engineer.


It is still a lot of money.


There's a software engineer at Google making $1.1M. That is all.


Nevermind, there's one making $1.5M. I just stopped looking too early.


I understand that highly compensated engineers get to solve unique challenges on grand scale, enjoy their jobs and are generally content with their professional positions.

But I just can't stop wishing for those engineers using their immense knowledge, authority and financial resources to bootstrap own tech companies and improve the diversity of the tech ecosystem. They don't need venture backing, they don't have pressing need to compromise and have the means to experiment with new business models, whether it is open source, open-* or social enterprises.


Pushing back a little bit, not everyone is cut out for the startup lifestyle and the problems they're solving may not be solvable without the scale of a place like Google. Rather than encourage them to get out, I'd love to see them publish more about what they're doing, what challenges they're tackling, and how they've gone about solving them. That information gets pushed into the general ecosystem and other people then can run with it and iterate on it quickly. I.E. Google publishing the big table papers that became HDFS/Hadoop and the entire open source "big data" ecosystem that spawned from that.


To add to your point, people who are trying for H-1Bs aren't necessarily the same people who can apply to YC on a whim and get a company off the ground in no time.


>everyone is cut out for the startup lifestyle

I would not wish startup lifestyle on anyone and I don't advocate for that. I may have been unclear, but the engineers in question are exactly the kind of people who don't need to take outside investments and are able to bootstrap (as in use their own funds and rely on immediate/eventual cashflow).

They would have the independence, resources and authority to create self-sustaining social enterprises ( maximizing social impact rather than profits for external shareholders.), rallying supporters around their mission.

Seeding acquired knowledge back into the ecosystem is an orthogonal/complementary activity and depends more on their employer. I am not sure we can encourage the engineers to do that. Also, while sharing innovation "blueprints" is definitely a net positive for the ecosystem, I think it disproportionally benefits the megacorps, while indies and middle-sized tech businesses are being slowly squeezed out. The web is becoming more centralized and I am wary of that trend.


> but the engineers in question are exactly the kind of people who don't need to take outside investments and are able to bootstrap (as in use their own funds and rely on immediate/eventual cashflow).

no, they're exactly the kind of person who wouldn't do a startup. you're thinking about this from the wrong end. risk averse people are generally the ones who make $1M/year at salaried jobs, exceptions like financial traders notwithstanding.

quite frankly if i could make $1M/year working for a megacorp i wouldn't have started my own company. these two things are mutually exclusive for me. why would i bust my ass if i could take 2-week vacations whenever i wanted and get paid $1M/year due to my elite skills?

in many respects, i started a company because i wasn't good enough to really 'make it' as an industry leading developer. i capped myself out as a consultant at about 150k/year. but i wanted more money.

many of my associates who also own businesses (tech or otherwise) also started because they were basically unemployable due to a variety of factors. issues with authority, massive egos, chip on their shoulder due to a previous fuck-over or failure, inability to kiss ass, did terribly in school, have a felony conviction from their misspent youth, etc.

these kinds of things aren't talked about in mainstream media but if you dig deep into an entrepreneur's past, you'll find this stuff a lot. normal, employable, salary-seeking people generally do not start companies of any kind.


I never saw it that way, but it does kinda make sense. Thanks for putting it down!


A bit more pushback on both points.

> I would not wish startup lifestyle on anyone and I don't advocate for that. I may have been unclear, but the engineers in question are exactly the kind of people who don't need to take outside investments and are able to bootstrap (as in use their own funds and rely on immediate/eventual cashflow).

The provide data is gross income, not net. Subtract federal and state taxes and you're at a more realistic figure for net income. Then subtract cost of living (i.e. property taxes, utilities) etc., and you're at a much lower figure for net expendable income than the numbers here. Your argument also assumes they're in a position personally to take on the associated risks of leaving stable employment and pursue a "moonshot" relying on savings and hopeful future cashflow. They might not be there. Also, running a business/organization is a far cry from slinging code and architecting and requires a different temperament, personality, and skill set. The individuals we're speaking of might be lacking in those or the desire to pursue such an activity. They might rather solve problems than deal with GAAP vs non-GAAP compliance issues, HR policies, marketing, or revenue generation. All things a corporate leader must address that an engineer solving problems might only ever have to address tangentially.

> They would have the independence, resources and authority to create self-sustaining social enterprises ( maximizing social impact rather than profits for external shareholders.), rallying supporters around their mission.

There are a lot of assumptions rolled into that sentence that I want to unpack.

First, what you're describing sounds incredibly similar to a non-profit and we're swimming in them. Why would engineers do a better job at building organizations that maximize social impact than people who spend their entire lives doing nothing but? The only comparative advantages I see are (possible) money and engineering acumen.

Second, how are we defining "social impact"? That's an incredibly nebulous term and one could argue that being part of a business generates social impact. Think of profit and revenue as an indicator of value provided. The more people buying and using my services generally implies I'm providing something of value. Ergo, pursuit of a profit and providing services people will expend money to pay for might be indicative of activities having a "social impact". People are willing to expend money for your solution to $PROBLEM over the possible alternatives indicating you're providing some value. As an engineer if I help the company solve a problem that lets them scale services to increase the number of people who can use it, offer new features, etc. I'm increasing the corporation's value proposition and providing a service to people to help solve their problems and meet their needs, I would argue that is a positive social impact.

Third, what is a "self-sustaining social enterprise"? Generally, I've heard those called corporations and they resemble normal every day businesses. What is different about your notion other than they're privately held? They'll still need to generate a profit in order to grow and increase revenues to meet growing payroll and expenses. I'm having a hard time understanding how you're seeing social impact as different from bringing in $$$ in exchange for providing a solution to problems.

The other assumption here I think we need to address is the notion that the individuals possess the charisma to rally people to the cause. You're describing someone with an incredibly charismatic personality that draws people in, that's a significant minority of the population. Why is that skill set attributed to engineers making big figures? Generally, I've found the opposite to be true, the charismatics and visionaries tend to not be engineers -- engineering is generally too rigid for the charismatics and the people stuff can be too squishy and shades of grey for the engineering types.

> Seeding acquired knowledge back into the ecosystem is an orthogonal/complementary activity and depends more on their employer. I am not sure we can encourage the engineers to do that. Also, while sharing innovation "blueprints" is definitely a net positive for the ecosystem, I think it disproportionally benefits the megacorps, while indies and middle-sized tech businesses are being slowly squeezed out. The web is becoming more centralized and I am wary of that trend.

The open source culture is booming compared to what it was a few years ago. More people are involved in OSS and more companies are pulling in OSS software to solve their problems. Is the system perfect? No. Will it ever be perfect? No. Is it better than what it has been or could have been? Heck yes.

How does it overly benefit megacorps? Yes, they might have the resources to pursue implementations faster, but a ton of the innovation on those systems is happening at the small to medium sized corporate level where the engineers have the freedom to innovate. I'm missing the link between the web being centralized and small to medium sized businesses getting squeezed. Generally, if there's a "squeeze" it's the small to mid sized businesses getting bought out by megacorps to fill needs, which, is usually an encouraged outcome for some of them because they aren't going to "get big" otherwise and scale up to the point where they have more than minute impact. Being inside megacorp gives them the opportunity to tap resources they would lack otherwise and get their tech into way more people's hands. I'll raise the case of the Storm processor. It started as a side project at Backtype and become Nathan Martz's full time gig when Twitter bought them out. It's one of the "big" open source data processors on the market and they're still innovating on it.

Ideally there's a balanced ecosystem. Big corporations have resources and manpower, but are prone to inertia and lack responsiveness in the face of new technology. Smaller corps lack resources and manpower but try to get an edge through creating new technology or adopting it, proving it out, and extending it. Big corps finally come around and either adopt the new tech along with the rest of the people in the ecosystem or buy out small corp to get the people who built or are extremely experienced with the tech/platform/framework. These people then get the resources + manpower to work on their projects. Eventually some of them get bored or hampered by the inertia, drop out of megacorp, and start or return another small corp and repeat the process. That's the trend I see playing out over and over again.

Yes, there ARE exceptions to the rule. I.E. LinkedIn + Kafka etc.


Thank you for a detailed response and an opportunity to debate this issue.

>you're at a much lower figure for net expendable income than the numbers here.

I am from a country where most people earn less than $500/month. Most Americans earn much less than the numbers here. This argumentation always seemed a bit absurd to me.

>and pursue a "moonshot" relying on savings and hopeful future cashflow

While bootstrapping, a top engineer/s is/are always able to do consulting while simultaneously building a product in the same/associated domain. EmberJS/Tilde.io is a great example.

>They might rather solve problems than deal with GAAP vs non-GAAP compliance issues, HR policies, marketing, or revenue generation

Sure. The preferred strategy is to devise solutions for making all of that easier. One solution is to be a CTO/founder while delegating administrative and some business matters to the CEO/CFO/CMO cofounder.

>..similar to a non-profit and we're swimming in them.

The number of tech non-profits is miniscule.

Let's look at the organizations who build popular database software. This is a quasi-random choice without any specific reason.

PG(non-profit), MySQL(for-profit, Oracle), MSSQL(for-profit, Microsoft), OracleDB(for-profit), MariaDB (non-profit), MongoDB(for-profit), RethinkDB(for-profit), Neo4j(for-profit), Complexible(for-profit), Apache Jena(non-profit), Apache Cassandra(non-profit), Couchbase(for-profit), Apache CouchDB(non-profit), OrientDB(for-profit). Those are major SQL, NoSQL, graph and semantic database products.

There are 10 for-profits and 5 non-profits (3 of them are under the stewarship of one entity -- Apache Foundation). The list can be certainly expanded and weighted by userbase for accuracy.

Another interesting domain would be collaboration tools for developers. The list of popular products is much shorter here: Github(for-profit), Bitbucket(Pivotal, for-profit), Gitlab(for-profit).

3 for-profits, 0 non-profits.

My argument is opposite of yours: the number of non-profits is too damn small.

>Why would engineers do a better job at building organizations that maximize social impact...

Technology is a great enabler. Companies with [1-5]0 employees have valuations measured in billions. Obviously, most of them had VC backing (money & network effects) but same principles can be applied to social impact.

>Second, how are we defining "social impact"?

Depends on the domain. Number of people who are able to afford expensive high-tech medical procedures, for example. This is an easy part.

>Think of profit and revenue as an indicator of value provided.... I would argue that is a positive social impact.

Not at all. Think BP, early days Microsoft, Oracle, Linkedin, MongoDB, etc, etc.

>what is a "self-sustaining social enterprise"?

Social enterprises are distinct type of entities with specific incorporation laws, see https://en.wikipedia.org/wiki/Benefit_corporation as one of the examples.

>..the charismatics and visionaries tend to not be engineers..

One doesn't have to be perceived as a charismatic leader by general population, respect and authority among engineers/programmers/IT professionals/geeks is quite enough. There are millions of these and they possess powerful technological skills to change the world for the better.

> How does it overly benefit megacorps?...

This discussion in the HN format would require too much citations, data and tangents to be efficient.

Income disparity, unemployment, lack of social mobility is increasing locally and globally. Technology, robotics and automation is one of the main reasons for these trends. Corporations are beneficiaries of these trends and won't do much to stop them. I would much prefer non-profits and social enterprises to constitute a bulk of tech businesses while for-profits would be a rare exception.


The pg essay quote seems to have gone out of fashion here a bit lately, but there's a perfect one for this occasion:

> So one way to build great software is to start your own startup. There are two problems with this, though. One is that in a startup you have to do so much besides write software. At Viaweb I considered myself lucky if I got to hack a quarter of the time. And the things I had to do the other three quarters of the time ranged from tedious to terrifying. I have a benchmark for this, because I once had to leave a board meeting to have some cavities filled. I remember sitting back in the dentist's chair, waiting for the drill, and feeling like I was on vacation.

http://www.paulgraham.com/hp.html


I quite disagree with the premise of this quote for the following reason. When commercial product is being built, external requirements and interests are imposed on founders and their aspirations to create interesting/innovative software will be subverted by the need to compromise. However, the essay addresses this problem.

The engineers have the opportunity not to take outside investments and run their companies as they see fit. I believe that those bootstrapped companies would bring more social value than startups who are restricted by the boards, investment agreements, valuations and competition in high risk, high yield markets.


You're missing the point: They still would have to run the business. Meaning, not do what they set out in life to do, namely engineering. Where they are, they can pretty much just focus on the problems at hand and do research. All of the other crap is done for them.


As someone who transitioned from 100% business/marketing side to 25% business and 75% development, actually I am not missing the point.

There are many upsides to running and owning a tech business and for some of the engineers those upsides can outweigh the negatives.


I've often had that feeling at the dentist. "Tedious to terrifying" just seems to be another way of saying "busy."


There's a whole class of problems that it only makes sense to devote resources to in the context of a large organization. Compiler, kernel, library work that makes everything a fraction of a percent more efficient is nearly unjustifiable for most small organizations, but of immense value to large companies because of the scale at which they operate.

If you happen to be interested in solving those problems, a large company can be a wonderful place to work.


You are absolutely correct and I would love to see a scalable, sustainable community-based solution where languages would be designed for the social value first and megacorps second. I would prefer this solution to offer financial incentives comparable to the corporate jobs.


I think the problem is even though it sounds like a lot, $1.5m/year is not really some kind of "Tony Stark"-like riches. Yes, they might be able to temporarily stop working and live off savings, but probably not for that long. And furthermore, $1.5m, which is really $750k after taxes, doesnt really pay a lot of salaries. Hard to have an engineering division on $750k.

Also, I know plenty of people who are doing exactly what you are saying. The reality though is bootstrapping a startup is unlikely to make it more of a success anyways. The VC game picks winners in both senses - that they fund 'winners' and that they create winners by their own influence.

So a bootstrapped startup cannot ignore the VC realm regardless, especially once it is time to come to market.

The era of a silent genius sitting in a garage building durable software for the ages is a huge myth alas.


The VC game picks winners in both senses - that they fund 'winners' and that they create winners by their own influence.

It's about 5% the first and 95% the second, sadly. The well-connected VCs pretty much know what the outcomes and the valuations are going to be in advance. The less connected ones get burned by adverse selection and lose money.

So a bootstrapped startup cannot ignore the VC realm regardless, especially once it is time to come to market.

That I'm not as sure about. You just have to bootstrap into something that VCs aren't likely to be interested in, and get enough of an early lead and enough good will that they can't just make a competitor out of nothing should their interests change. It's quite possible, as far as I've seen. Bootstrapping is far harder than it should be (in part, because the American middle class is dying and no one has any fucking money) but I don't think that the VCs are the cause of that.


not sure why you've been downboated, perhaps people disagree on the numbers, but the basic fact is, in my opinion, not disputable.

VCs dont just pick winners, they CREATE winners. How much of an effect this is debatable but it certainly is not a small effect.


>It's about 5% the first and 95% the second, sadly.

I do agree and believe most VCs successes to be self-fulfilling prophecies.


Speaking as an above average engineer, but not one of the super spectacular, I envy the work-life balance that many of these get to achieve, avoiding a lot of the slavish work tech startups often put them through. Even if they funded out of pocket, it would be a long hours, market pressured situation, where enough VC or big company money can still bury you despite brilliance.

I'm honestly not surprised some choose the good life at a big company - the world often doesn't treat such skilled people well.


If youre well paid to solve exactly the kind of problems youre interested in working on, its easy to see why some would prefer working at Google / Facebook. Most technical people I know who have left to start their own business have shared that they now spend most of their time building the organization rather than creating technology. Some appear to enjoy this organizational hacking, while others have a tinge of regret about it.


1.1M in Pittsburg is more than 1.5M in Mountain View though, so I'd stand by the first one being worth more


Well, that really depends.

Once you pay off your house, adjusting for cost of living doesn't make as much sense, especially for areas where rent or mortgage payments makes up a huge percentage of expenses.



There is a "Data Scientist" at AirBnb with a salary of $1.35M.


Most of the other Data Scientists are making ~$135k, maybe just a typo or parsing error resulting in an extra zero?


With a location of "Pittsburg, PA"


Would love to have a dump of the data: Every time people scrape these data, they have to clean it again and again.


For all the nay sayers, this information provided is for the good of all.

If a company declares they can't find talent here in the states and want to hire from outside this information helps:

- Makes sure the company doesn't just go hunting for cheaper labor overseas and bring people here for 40k

- Allows the international employee to view salaries to see if they're getting a fair offer

- Helps U.S. Citizens keep there job/pay because the company can't go higher other people for 40k.


1) To prevent hiring people "below 40k" you just make it so you dont pass H1B's for "below 40k".

2) International employee does not normally know something like this exists, and I assure you the Government website is not made with such a goal in mind. Not to mention that even in that case the salaries provided here have all kinds of inexact data (because it doesnt show stock compensation)

3) Any country can make the immigration policy they desire. Having this information public denotes a second-class citizen measure. This would not be acceptable if, for example, employees from other states in the US had to do this to "protect the state's jobs and make sure everyone gets a fair salary".


"Having this information public denotes a second-class citizen measure"

Actually it denotes "not being a citizen at all"? It's got nothing to do with being second class. If you are here on an H1B, you're not a citizen.

A google search shows: "The H1B visa classification permits a foreign national to work in the United States for a temporary period".

You can debate whether the information on this site should be public or not, but it is not reasonable to think that a foreign national should have all the same rights as a citizen.

I would not expect to have all the rights of a citizen if I went to work in a foreign country either.


"I would not expect to have all the rights of a citizen if I went to work in a foreign country either."

Your expectations are accurate.


Some really strange application from IBM :P

H-1B Senior Software Engineer $7,278,870,000 Year $63,294 Philadelphia, PA Unknown 1 Withdrawn 15-1131


According to this site, my company has LCAs with salary 12% to 26% MORE than what I'm making for the same title! I didn't know I was getting screwed like this. I was happy with my job until I saw this site today.

I'm not sure what to do now. Should I tell my boss about this and ask for a big raise this year? How would they react if I showed this data to them and demanded that my salary at least match these numbers?


Sorry for that. I don't think anyone can predict how your boss would react to such demand.

That said, it might be that those colleagues making more than you do simply offer more value to the company, or were better negotiators when discussing their job offers.


The salary information is potentially very useful in negotiations ..


I'd be a little careful with that as it may end up anchoring your salary.

The prevailing wage may not reflect the actual prevailing wage (which is why the H1-B visa is attractive to some employers) and if companies aren't paying much more than the inaccurately low prevailing wage then you may end up short changing yourself.


The site shows average salary, but it would be nice to see Median salary. In fact, it might be best to split the salaries into different quartiles.


Or get a full histogram!



unfortunately that site hasn't been updated w/ 2014 or 2015 data


Check out Wal Mart during year 2002. Either there is some interesting story or the data has a problem. Average salary goes up by about 50 times before dropping back down by about as much the next year.

Edit: Being there are so many Wal Marts, I mean this one. http://data.jobsintech.io/companies/wal-mart-associates-inc



The other Wal Mart. http://data.jobsintech.io/companies/wal-mart-associates-inc

Actually, there are quite a few Wal Marts.


Pharmacist $79,040 Pharmacist $7,904,000

Looks to be just due to errors in decimal place.


Weird seeing yourself listed on a site like this! I imagine my co-workers could find me if they were so motivated.

(this isn't a criticism of the site - just an interesting side effect of this kind of data transparency)


Yeah, it was odd to see myself listed in there, too.

Although, at the moment, anyone who's interested enough to look at the H1B notice posted discretely on the wall of the kitchen at my office will see my salary, since it's my visa being transferred to our new parent company.

It was a strange feeling seeing the job details and the salary listed on the communal kitchen wall, and realising it was my job and my salary.


Last time I said this I was down voted. This kinda transparency needs to be there for citizens and residents too who don't have to go through LCA process.


I've always been in favour of transparency around salary and compensation. So, I'm not hugely perturbed by this.

It's a bit strange when it's done piecemeal, though, and without your knowledge - I jokingly asked my boss "is that my job listed on the wall, or are you hiring someone to replace me?"


Absolutely, but on a personal level it leads to some inequality - my co-workers know how much I earn, but I have no idea how much they earn.


When searching for "google", one can find few interestingly named companies, like "google iniiii iiiiiiiiiiiiii": http://data.jobsintech.io/search?utf8=%E2%9C%93&employer=goo...

I wonder what those are about.


that's the next big thing


Interesting to see the salary ranges at my employer. With similar titles, the range is quite large, even within the same physical location. And modifiers like "senior" or "principal" don't appear to correlate with the listed salary.


I would love to be able to query the whole dataset at once, instead of company by company.

For example, I'd like to see everything in $city, and then be able to sort that by salary or job description.


It's in the pipeline :)


Really interesting project. If you look at small but relatively well known companies, eg Magic Leap[1], it's possible to get some real insight in to where they're focusing their efforts.

[1] http://data.jobsintech.io/companies/magic-leap-inc/2015


It's a really clean implementation. However, it took me a while to understand what the website was all about (wasn't sure what LCAs are, etc.) One improvement could be making the messaging more accessible and less technical.

We are interested in an API and would love to hear your plans about it. Will shoot you an email :)


Thanks. Can you shoot me an email theo@jobsintech.io ?


Very cool site! I found similar data a while back, but nowhere near this organized/pretty to look at. It helped me make the decision of NOT taking a job that I had a hunch was a lowball offer.

Does anyone know if companies are allowed to include the bonus $ in the 'salary' figure in these LCAs?



Thanks, this is a very good job so far. I do have one question, though.

What is your reasoning for having the upward-facing arrow /\ serve as the descending sort and downward-facing \/ as ascending? It feels backwards to me and a bit counter intuitive.


The asc/desc triangles bugs me too. Arrows would indicate order (⬆︎ = rising, lowest to highest). Triangles aren't arrows and should IMO be the reversed, like a pyramid (▲ = ascending, smallest on top). I just noticed HN uses triangles as arrows for upvotes, whereas Reddit has the arrows.


Good to know I'm not alone in this reasoning. I've seen the flip-flopped behavior on other sites, and it is a pet peeve of mine.


Websites are one thing. It really gets annoying when different applications in an operating system "can't agree" on what to use. I've seen both in Debian/Gnome apps, but that's kind of to be expected from (F)OSS software as rules for UI are less strict/not very well defined.


I was thinking from the top salaries to the bottom.


I was confused by this as well, usually it's the other way around. I think regular reasoning is from largest (widest part of arrow) to smallest.

Thanks for the site though :)


You need to dedupe the companies better. I just searched for "LinkedIn", and found several matches:

    1 	linkedin, ltd. 	
    2 	linkedin, inc. 	
    3 	linkedin, corp. 	
    4 	linkedin ltd, inc. 	
    5 	linkedin corporation


So many ways to misspell "microsoft"!


And "ibm corporation", too!


Minor bug: the Home tab stays highlighted regardless of which page you are on: http://data.jobsintech.io/companies.

Good job otherwise!


Thanks! Will fix asap.


For all the H-1B is "cheap labor complains" the fact that this information is public is horrible. Its a very asymetric relationship with your colleagues. If this were really meant to be about transparent controls, it should be tied to the wages of the other employees.

Cant believe there isnt a single comment about privacy issues regarding this here.

EDIT: after scrolling a bit I did find some. Kudos to that conversation, though this has to be off-putting. The OP is also playing with people's information. He hasnt put his next to it, for example. Not that it would have been enough.


People who know me, worked with me, ... can find me easily in this database and checkout how much I was paid.


I would find it sensitive to literally 100ks of people to make an effort to anonymize the data. If you don't see the ethical hazard of something like this, who is making the site, who is going to? The legislators that already passed the law?


This information is all publicly available already. Nothing new other than the interface.


If you think a site like this has no effect because the information is already there, you should think its absolutely useless.

But its not, because it does have an effect. And that effect is negative for a lot of people who are being treated different for their nationality.


What's being used on the front-end to display the tables and pages and the back-end to store all the info?

Time required to create?

Other fun details?

(I ask because I have another dataset I want to work on recreating online in a searchable manner)


Front End: twitter bootstrap. Back End: RoR+postgreSQL I roughly spent a week on it. At first I was cleaning and moving some data around with RoR but it didn't seemed right to me. Then I started poking around with postgreSQL and once you get a hold of it, it's super powerful and does most of the job. Use indexes... on fairly large datasets it really makes a difference.


Thank you for doing this! Any tips on how I can learn and better at using twitter bootstrap? So far I've only figured out how to load the basic template.


Let me know if you need a helping hand. :)


Even though this is anonymous, at big companies like Apple, you can see executive salaries with specific enough job titles that it's easy with a little LinkedIn sleuthing to know the person's identity. http://data.jobsintech.io/companies/apple-inc/2013?order_sal... - you know the person is a foreigner, and you know about when they began their job.


Here are some other interesting queries:

Associate and Analysts at Kleiner Perkins http://data.jobsintech.io/companies/kleiner-perkins-caufield...

Partners at Sequoia http://data.jobsintech.io/companies/sequoia-capital-operatio...


Hmm, was hoping this would show the denied percentage, but the "denied" column doesn't corresponds to reality, not sure what it's supposed to represent.


this only shows LCA (Labor Condition Application) data. H1B can still be rejected/denied much later in the process.


Also LCAs are filled when a foreigner under H-1B visa moves from a company to another for example.


Does anyone know how accurate employers are legally required to be with these? Is this just base salary, or are they allowed to include cash bonuses, stock bonuses, etc?


This would be the base salary.


Nicely made. It would be helpful to view the information grouped by job title, with summary statistics on salary across all employers, possibly also with standardized numbers that include adjustments for differences in regional cost of living for the various worksite locations (via either BLS numbers or whatever wolfram alpha uses to get equivalent salaries for other cities).

It's also a little depressing to see salaries this high (I'm at a non-profit).


It's a little exciting to see salaries this high (I'm at a for-profit).


Do these numbers include bonus/stock or is it just base?


when you fill out the LCA, you only supply base - not total compensation


Wow, immigrants have no privacy in this country. Imagine, if US citizen salaries were made public like this. How would that make you feel?


Actually, I worked for a the UC system where all salaries were made public in an easily searchable database. The sky didn't fall.

My only real problem with it is that I feel it puts the UC system at an unfair disadvantage. Imagine if you could look up salaries for your competitor's workers, but they couldn't look up yours? This is essentially the situation for UC relative to a lot of private universities, even though massive amounts of government funding flow into private universities (with a tax exemption on endowment growth, too).

I actually think that all companies that make use government subsidies program should be required to disclose employee salaries the same way that state government do. And I absolutely do think that the H1B program should count as a government subsidy (Milton Friedman recognized it as a subsidy, because it was changing the rules for only a certain segment of the economy).

Overall, I actually think it would be beneficial to publish all salaries, and I really don't think it would cause the kind of problems people worry about. It would also real even the playing field between employers and employees.

Think about it this way - all real estate transactions are made public. But imagine if you were bidding on a house, but only the real estate agents actually knew what every house in your neighborhood had sold for. You were forced to go by little bits of gossip here and there. You'd be at a terrible disadvantage!


I worked for the State of Virginia a some years ago, and at least two years running the local paper did a FOIA request for everyone's salaries and then published anyone earning above the median in a searchable database. ( http://www.richmond.com/data-center/salaries-virginia-state-... )

Many of my coworkers were upset, but the only real practical result I saw was that some people that were underpaid relative to their peers had evidence of it, and those that were overpaid relative to their peers were held to more responsibilities by others.

On the whole my experience was positive with it. (That said, your point that we have varying levels of "justice" is completely valid).


I am an advocate for open salaries, because it eliminates or helps to eliminate the information asymmetry between employers and employees, that absolutely favors the former. Partially giving this information just deepends the information asymmetry between colleagues.


My wife used to work at a public university and their salaries were publicly available on the university website.

You could literally search "EMPLOYEE NAME salary" on their site and find any employee's annual income.


In Nordic countries tax records are public so you can check anyone's income if you know the name and birth date.


Public employees salaries are public records in the US. Many local news publications will publish a searchable list.


The salaries of private company employees are being listed here


This needs to be transparent like it or not. These companies are arguing that there isn't any talent for them to hire so they have to get someone from overseas. If this information wasn't publicly available they would be paying some of these guys 9 dollars an hour.


That is senseless, you dont need citizen control to check that the salaries are prevailing wages, that is a job that can be done by the government with multiple different agencies at a very low level of cost or risk. Also, it doesnt make sense because it doesnt include US employees salary of the same companies which is what you need to make that fair/unfair comparison. Even worse, the immigrants themselves can't make that check, which is the most important one you need.

The fact that this information is available to citizens is a purely exclusion measure, that makes clear that immigrant workers don't have the same rights as US citizens.


The fact that this information is available to citizens is a purely exclusion measure,

No it's not. The records in question aren't public because they involve foreign workers. They are public because the records involve a program run by the government.


So are the records of the incomes of citizens. That's how the IRS decides how much you owe. Yet those are not published.


The IRS records of foreign workers aren't published either.


Their salaries are. That's the whole point of this submission.


You obviously don't know anything about the US government. Without transparency there is corruption. Why do you think everyone is upset over the Trans-Pacific Partnership deal being done in secret? The US government can not be trusted.

The most recent example of this: http://www.washingtontimes.com/news/2015/may/19/pentagon-was...


The only information in this case relevant to check is the salary itself. You can avoid the company name entirely, have only title and location, and get that information as well, and that protects the identity of foreign workers considerably.


Your question concerned US citizens not public / private employment.


Of course, hypocrisy runs rampant for this community. Privacy is important... as long as it's our privacy, not their privacy.


This not an 'immigrant' thing, it is a H1B requirement. When I initially began working in the US under Trad NAFTA there was no such requirement, but when I changed to a permanent track the job had to be posted. It is part of the process to prove that there is not a qualified person already available to fill the position. The whole underlying proposition of H1B is there is a shortage, disclosing the salary is just part of the process of proving it, which falls to the company making the petition. Really, I never gave it a second thought.


It would be a step in the right direction.


Well, I suppose that if the employer and employee couldn't normally work together but had entered into a special arrangement with the government dictating (among other things) that their salary be made public in exchange for being able to employ/be employed, I'd be just fine with it.

The situation is not remotely analogous.


Nicely done. This is useful to gauge salary levels in your current company and also what companies in similar domain are paying in the area.

There is another website http://dolstats.com/ that gathers data from PERM applications. I've found it useful for comparing salaries when looking for new job.


The USCIS publishes this data in an excel sheet format. You can find the raw data here http://www.foreignlaborcert.doleta.gov/performancedata.cfm . I made use of this data to negotiate salary and came out successful.


Or you can use https://wageoffer.com


This site is nicely done and very clean. I used to get similar information from myvisajobs.com before. Well done :)


Are L-1 visa salaries publicly available? I know some Engineers who've come in on L-1 visas.


How is it, that in the US people earn between 20k and 200k this easily? It's just insane...


The cost of living is extremely high in many of the places where people earn these wages.


Not 10x higher, though. Also cost of living is now always matched by median income. For instance, in NYC the median household income is 50k - the same as the rest of the nation.


It strikes me as odd that Microsoft is second[1] - being ahead of companies which specialise exactly on acquiring workforce from outside US.

1: http://data.jobsintech.io/companies


I recently started toying with the idea of moving to the US so this seems immensely useful.

By the way, does anyone know a good blog post or something on the process of applying for H1B jobs? I don't have a clear picture on how the whole thing works at this pont.


H1B here. Two small caveats with these data sets.

1. They may be incomplete. My position (2010) does not show up at all here, though it does on one of the other similar salary trackers.

2. They do not reflect raises. My salary is ~14k more than my starting salary over 3 years.


That's cool. There's a bunch of site that have this data. http://salarytalk.org/h1bwage has similar information on the data set.


Well damn. This really tells you how hot the market is and how much information asymmetry there is in job negotiation: it is interesting to see the variance in each company at the same prevailing wage.


Are we overrunning the server? Seems like it's gotten slow.

At first indication the #s seem pretty accurate for where I've worked, though they only capture base salary. (No bonus, equity, etc)


I didn't know salaries of visa holders are public information!


Shameless plugin. Also checkout http://visadoor.com, which provides the same information and more.


Is there a way to remove "withdrawn" applications from averages? There are a lot of decimal point errors there, which skew the average way up.


How would you distinguish this from visadoor.com? What does it do better?

(Not trying to challenge or insult--trying to learn.)


I have a lot more data(millions more) and a really clean&easy UI :)


Someone at IBM: Senior Software Engineer $7,278,870,000 Philadelphia!! Probably some lucrative patent.


Much more likely just bad data (astronomically more likely)


I agree with you. I was like whaaaat..


TN Visas for Canadians and Mexicans seem to be missing, even though H1B1s and E3s are shown.


Yeah, they are not the same class.


Search for "Amazon" gives me 87 different companies



very interesting now i know how much my manger makes !! data is scary !!


Did anyone check out netflix? You can really see salaries spike in 2012/2013 time frame.

Is this due to the massive wave of VC-fund companies now competing for talent?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: