The finding that most surprised me was that the mean salary for an engineer without a college degree is only $3k (~2%) less than for those with one; this gap is much smaller than in the labor market as a whole. One explanation is that CS really is a field where educational signaling doesn't (or at least needn't) matter as much as in other industries - we recently discussed this with Bryan Caplan over on our blog (https://triplebyte.com/blog/bryan-caplan-interview). I'm self-taught and don't have a CS degree, but I do have a college degree which still opens doors. I'd be curious to hear from other developers without a formal background on this.
Boot-camp grads average $19k less - but $130k is still quite a bit higher than I've seen bootcamps advertising. Could this be indicating that they're at a disadvantage in the normal hiring process just for signaling reasons?
The sample set of people using TripleByte is not going to be remotely representative of the market for developers as a whole. Candidates with stronger resumes are not going to be using third party recruiters to spray their resume around; they are going to be applying directly to the companies they want to work at or getting headhunted by the companies themselves.
- TripleByte imposes a cost (having to go through their process) in exchange for getting to signal programming ability, a prerequisite to getting in the door (getting an interview)
- People with strong credentials don't need to incur that cost, so they probably won't do so.
This means TripleByte's pool probably doesn't have any Stanford CS grads who are looking to leave their Google job. But these are also the people who command the highest salaries. Let's say it chops off the top x-percentile of the market, for some reasonable definition of x (10? 15?)
Then, TripleByte's screening process probably also chops off the bottom y-percentile, because those are people who can't actually pass the screening.
Once you restrict the range like that, and also make paper credentials less relevant because there's an alternate signal available, of course the rates are going to be compressed.
Why are those mutually exclusive? I like to think I have a decent resume and would apply to companies directly but if I can get more offers using Triplebyte why wouldnt I? Isn't their draw that you don't have to interview at individual companies but can use a streamlined application?
After comparing to a much smaller set of broad-market resumes, it was clear that you’re right: people using online job finding tools are actually shut out, for some reason, from conventional job channels where the intuitive rules apply.
But yes, obviously there’s something wrong with the data if the salary gap observed was so small. I generally like TripleByte and its writing, but it seems like they really got something 200% wrong here.
We are a small team of 5 expecting to double in 2019, all from experienced folks extracted from big companies. We budgeted $260 each and our investors didn't bat an eye.
This is an interest statement.
It is interesting, because the data I am seeing is that apparently Triple Byte is able to give people quite successful results. (Yes, 150k base salary is pretty good).
But your statement seems to imply that candidates that use them are apparently below average or something. And yet even though they arent as good candidates, they are apparently able to give people really good results, despite that.
Is my barometer for tech salaries so off, that apparently people believe that 150k base isn't something that a "stronger" candidate might receive? Yes, I've heard of some high salaries, but I still wouldn't call this something to scoff at
I'm a high school drop out with no diploma. It's never come up during any hiring process in the 17 years I've been doing this.
That said, I've only had two long term jobs. I spent years doing contract work, and I mostly got contracts by word of mouth.
I've been on a number of hiring teams, mostly at Puppet. I was also a hiring manager at Puppet for a while. Education never came up once — I don't remember even talking about it in interview training.
MS is the exception to that rule.
I work in New Jersey, as an Enterprise App Full-stack dev with 0 college experience.
I started as a developer 4 years ago @15/hr, and I recently breached the 6 figure mark. I've told my coworkers that I have a fire lit by a sense of inadequacy. I've always been behind classically trained developers, that is what keeps me pushing forward. Always playing 'catch-up'.
Now that I'm involved in my company's interview process, a few thoughts on what having a college degree does for our offers;
1. Because we hire through an agency, we already have a single layer of vetting that helps remove unqualified persons (both college-level and not), which ensure we have a decent, homogeneous pool to conduct face to face interviews with.
2. Whether we hire a candidate with or without a college background, our the offer range isn't enormous. If we have an offer in mind, (say 75k), having a college degree doesn't automatically grant you the high end of our scale. I can't think of a single instance where we cared about their degree once they were in the Face-to-face. Their performance in the interview dictates their offer, and we aren't asking questions like 'How do you implement bubble sort'. We ask some hard skill questions, sure, but we also ask just as many communication and general problem solving ones as well.
3. The range of starting pay is really small for us (think 75k mid, 70k min, 80k max). So that 3% gap makes a lot of sense.
To summarize, a degree may put you in a position to interview, but the range you get paid is largely depending on a wide array of skills, some of which are unrelated to development. Missing some of the hard-skills won't disqualify you for a position as much as missing the communication skills will (for us).
Total Comp figures are important, and differences in effective take home pay will be hidden if you only list salaries.
and yes, it is underwhelming that we still don't know "how much software engineers actually make"
The only thing I feel like I missed out on compared to a CS grad is a better intuition for big O. I just don't have any use for it in real life, so it never really clicked. Good benchmarking tools are all I've ever needed. But it would be nice for my own intellectual gratification.
I have found the biggest discriminator on hiring salary and hiring generally is due to approach to problem. If you are limited to a single convention, such as OOP, or are limited to certain tools/frameworks businesses are less eager to hire you. In theory that makes sense in that you are more valuable if you can provide an original direct solution to a problem with greater ease. In practice it doesn’t make sense because those limitations tat prove to be a negative bias during hiring tend to be the reality internally.
Care to share the city? Or region?
I've also seen people graduate who couldn't do anything, so a CS degree is not proof that someone can do the job, and not having it is not proof that someone can't. I think university education is still extremely valuable, and I'm glad I've had it, but more for what I learned than for the piece of paper.
That said, salaries in Amsterdam, with or without degree, are far lower than the $130k to $150k described here.
I wouldn't expect Triplebyte to be able to observe this, because it only works with candidates who are US citizens, permanent residents, residents of Canada, Mexico, Singapore, Chile, Australia, or, have an existing H-1B visa (per the Candidate FAQ: https://triplebyte.com/candidate_faq).
Are these base salaries or total compensation?
Can you share what the distribution of companies you source to is? Ie are they mostly early stage startups or do they include many large post ipo tech companies?
Triplebyte works with over 400 companies, including Apple, Dropbox, Instacart, and numerous early stage companies (including many YC companies). The Open Positions section on the salary tool page should give a pretty good feel for the kinds of specific roles and companies that work with us
I think I may have found a tiny bug: the “all Triplebyte” dropdown seems to always reset the previous dropdown to “all company sizes” and visa versa so it’s impossible to have both selected to something other than the first options. Perhaps this is because data doesn’t exist for both? Maybe it’s just a slightly confusing UX.
It’s such a thoughtfully considered tool that I thought you’d appreciate my considered feedback.
Thanks!
Why are you mistakenly confusing education for education signalling?
There are many types of skills required to be an effective software engineer. Most of them have nothing to do with mastery of a programming language or technical tool, and have zero connection to solving coding puzzles in a short timeframe or memorizing answers to classic systems design questions.
You need to be a skilled writer and researcher to deduce business use cases and write effective summaries, presentations or user documentation.
You need appreciation for potentially many other knowledge domains, from legal topics & security to applied sciences. Having basic coursework in calculus, chemistry, physics, rhetoric, history & civics, etc., are crucially important in business settings.
It seems so tone deaf to me to baldly state that education signalling is a factor here, as if signalling was the phenomenon (it’s not).
Education (as opposed to education signalling) is a very valuable thing, and certainly fosters more effective engineers by a landslide.
The popularity of hiring from bootcamps or non-traditional engineering backgrounds is a commoditization issue, meant to suppress wages from growing as labor productivity creates dramatically greater returns.
There’s no shortage of engineers... there’s a shortage of “cheap” engineers (and yes, $130k is a cheap price for these types of hires).
My open source/side project experience has opened doors though I'm sure.
If you take their numbers that works out to $350-$400k.
A director at Google wouldn't be making much more than $200k in base salary, but their total comp can easily exceed $1M if you count stocks.
Yes. Most boot camps are outright scams and I know that I (along with everyone I know) has never hired a bootcamp grad. They usually cannot write fizzbuzz.
Some anecdata on the college grad note: most of the best engineers I've worked with did not study CS in college, though they all did graduate from college.
I'm a little suspicious because the median and mean are equal for several of the larger subsets (years of experience). I clicked several combinations to filter by and the difference between the median and mean was always low-mid single digits.
Without insight into total compensation, these numbers are only minimally useful.
I worry that some really skilled developers may be looking at these stats and not realize what they are giving up by accepting a job with the average salaries reported here.
It's riskier than cash but definitely real compensation and not the monopoly money that options end up becoming.
Of course, you'd be better off going to FANG or a (soon-to-IPO) unicorn where you'd make double these amounts with RSUs factored in. But if you can get in there, you probably don't need to go through TripleByte.
Then again, TripleByte also created https://www.levels.fyi/comp.html so maybe I am just being cynical.
Worked with lots of recruiters, personal network etc., felt like I really got a sense of what the salary landscape is here. Some facts:
- I was leaving a job at a hot unicorn startup that paid $125K base and I was told I was one of the highest paid engineers on staff (20+ engineers)
- I received 2 offers during this job search (was very selective else could have gotten many, many more)
- First offer: IOS developer at very large consulting firm for $120K base (with possibility of small bonus).
- Second offer: IOS developer position that was comparable in interest to first. Decided to "shoot for the moon" and ask for $145K. They agreed, and that's where I'm (happily) working now.
According to this study, SW engineers with my level of experience in these cities are making $181K on average! From what I could see that's just not available here to the rank and file, "average" experienced sw engineer. That salary would be more like top-of-range here.
Am I doing something wrong? Is Boston really that different than Seattle/NYC/SF?
For example, company A offered 100k, B offered 120k, and C offered 130k. You want company A. You say to company B that company C offered 130k and you're interested but not sure with that diff. They up to 135. You say to company C, hey, company B offered 135, can you help me lower the diff? They up to 140. Then you go to company A, say you'd love to, but you have offers for 135 and 140, so if they could do anything to lower the diff, great. They offer 130. At this point, you have three offers, all higher than you began with and can make a choice.
And if you want to count my pre-teen years coding BASIC games, total experience jumps significantly!
That seems mostly to be a supply/demand thing (which is tangled up with local culture norms, too, in complicated ways). There is much, much less VC funding in Boston than in SF and NYC. But a great supply of engineers because of the universities. And there's no equivalent to Microsoft+Amazon (Seattle) and Wall Street (NYC) pushing up both demand and top-end base salaries for experienced engineers.
Also, you mention that your search was 18 months ago. In SF (and I think Seattle and NYC, too) the job market is crazily engineer-favorable right now. I'd bet there's been 10% inflation in average base senior engineer salaries in SF in the last 18 months. That would match your $145k against ~$165k (rather than $181k).
As for funding: https://techcrunch.com/2018/08/04/boston-area-startups-are-o...
I don't think Boston is as much as a slouch as you think, in fact it's is pretty well positioned for the long term in regards to overall industry diversity and pretty respectable funding (though not as dominated by software like SF might be)
Salaries are probably lower because overall COL has traditionally been lower here (though housing sure isn't far behind anymore). Most people I know working for the big companies are getting ~$250k total comp (8-10 years exp). Other large, if more under the radar, companies are not too far behind.
You're not doing anything wrong. I just don't think Boston pays these kind of salaries (yet). We've got a burgeoning tech scene here but compensation isn't at the same level. When I was looking to relocate back to Denver a few years ago, Boston salaries were much higher than what I was seeing out there.
You could absolutely command much more than that. I've got 20 yoe, was VP level at a smallish startup in Cambridge (35 employees, about 20 of which reported to me) that was acquired. When it was time to move on, nearly any conversation with recruiters started with "total comp" expectations. I had that discussion with about 10 recruiters, and only 1 outright said no thanks when I threw out an expected total compensation of $350k.
For public companies that meant base, bonus, and RSU. For private companies that meant base, bonus, and paper money.
I eventually interviewed with two companies (1 private, 1 public) and received offers from both. Both offered compensation far, far greater than $180k.
On the west coast you'll usually get paid a significantly higher salary in Seattle/SF than most of the other cities with any tech presence (e.g. Portland, LA, San Diego, etc).
Salaries are tied to cost of labor/employment, which varies by city. It is generally correlated with cost of living, though (otherwise people move away), so it balances slowly minus network effect adjustments.
FWIW, I thought the SF numbers in that article were low for mainstream companies, but high for startups.
Personally out of school I got 85, but got bumped to 110 after 1 year and 120 after 2.
Maybe they’re including 495/NH jobs bringing it down? There definitely is a difference between jobs in Boston/Cambridge, those on 128, and those in the outer burbs.
1) SF/NY/Seattle salaries are skewed by really rich companies 2) Yes, there a big uptick as it's more competitive, FYI cost of living is much more. 3) The salary I think has a selection/sample problem that probably over estimates.
My bet is $150K for an experienced dev at a normal company in Boston is probably somewhere in the league of normal.
Consider that salaries are not as an efficient a market as we might imagine. If Boston area companies can get away with paying $140 instead of $180 they will! Most people don't move across the country for a pay raise. Worker mobility in the US is down over the last 30 year (weirdly).
SF has a lot of people moving there which creates a different kind of frothy market.
Boston is established and so everything gets established, including salaries.
Consider the situation: imagine if you had a company making ok profit, with 50 Engineers in Beantown. Do you think you could all of a sudden get your Eng. to the 'next level' by paying SF rates?
So you're staffing costs are way up, but is productivity? Surely, you might be able to bring on the best hires, but will that make all the difference?
It's such a big bet, and the inclination is always to make (or save) money 'now'.
It's a little bit like the 'open vs. closed office' calculation. The CFO can make a direct and measurable compelling case for open office: it's 20% cheaper. Those in favour of a nicer office can't provide the hard numbers on how much the company would increase productivity.
I think salary differentials are a crazy interesting subject and suspect there are a lot of weird and interesting things going on.
I'm from Canada, where any good developer can double their salary by moving to a choice job in the US. Why the hell would young talent stay? Which implies, how the hell can Canadian companies even remotely compete on building great companies of the top tier talent leaves?
Obviously this depends a lot on a companies specific need for hyper-top tier talent, those that don't would maybe be better off out of the Valley.
I also wonder a lot on what would happen if a well funded Canadian startup actually started paying super great salaries. Sure, they'd be able to get the best talent that comes through the door ... but then there's the other paradox: most Canadian cities are not destinations! Toronto, Montreal, Vancouver are regular cities, if you start a company there, most of your applicants will be local. It stands that local talent may not be all that spectacular (good but not great) thereby not justifying really big salaries. And can you really have 'regular salary' for the 'regular, local talent' and then inflated salaries for the international hot talent? That might be hard!
So SF, NY and Seattle have the advantage of being destinations, i.e. cities where people are willing to move, meaning that their quest for talent is really a not a function of the locals, but the top tier of a much broader pool of talent.
Certainly for some people SF or Seattle have a lot more to offer than just career opportunities and not everyone is going to like the Vancouver lifestyle but in my experience the primary draw of SF is money and career not desire to live in the cities for other reasons. I think the story might be a little different for NY where lifestyle might be as big a draw for many people as career opportunities.
Despite the fact that a polygot dev. with tons of experience is always good to have around, and you need some weighing in on big decisions ... most dev is just dev - and a competent mid-tier developer with good habits can do the job in most cases.
People going into software should plan accordingly. You get great wages to start, but you need to save some of those for a potentially risky mid-career transition where you specialize in a particular subfield.
There's a lot of managerial things you continue to learn, but that requires you to be in a position with enough political capital to actually stop history from repeating itself. It's only useful knowing that Project X will fail, if you can convince people to instead do the more successful X++. I'd expect job title to be a much better indicator here: having 20 years of experience doesn't mean you have the aptitude or inclination to succeed at office politics.
I think Triplebyte is stuck between a few incentives that aren't necessarily aligned with developers, and I'm not sure how to decide how I feel about their contributions overall to the job market for software.
I absolutely love their outreach and data like this where they try to empower developers more.
My problem I have is that they entered the market using the same exact worn-out old methods that everyone has been trying to use to hire for software forever. They do a timed coding challenge, they ask we-need-to-hear-the-right-answer quiz questions about both algorithms, and topic-specific areas for web/mobile/back end.
All they've done is dress everything up in fancy clothes and used an interface UX tuned towards developer's sensibilities. They absolutely bombard all major programming related internet sites/communities with their stupid click-bait "only 2% of green-skinned back end browser developers living at least 100ft above sea level get this question right!!!!111!1!1one1!" ads. Their initial, multiple-choice "gotcha!" quiz should be passed by a second year CS (or any eng.) student with a tiny amount of thought, but it makes developers feel like they've already just "beat" someone else out and "accomplished" something and it plays a psychological game with them to get them to commit to the extended live interview. (It worked on me!)
Do they have data about how their candidates perform after hire? 2 yr performance reviews? 2 yr turnover rates? How do they market to the recruiters who are their true "customers"?
To me their product seems like something that hiring managers and in-house recruiters can use to remove liability for bad hires. Has anyone ever worked with them through recruiting to see if the value proposition is there for eliminating bad phone screens? I don't think initial phone screens are terribly difficult to do for a senior engineer with experience, to root out red flags. Is it a better value prop on your expensive on-site interview slots to trust Triplebyte? Or your own internal senior engineers/hiring managers?
I'm not saying that Triplebyte has direclty nefarious intentions or anything like that. I just am no sure I see how their incentives are aligned with me as a developer, in quite the same way they market as ("We are god's gift to developers").
Software engineering hiring, and Triplebyte as a company, at least think about, and make an attempt to evaluate these concerns. Have you ever seen how non-technical people (sales/marketing/finance) are evaluated? I'd argue Triplebyte at least moves the needle a bit versus in other fields.
Honestly, I'm not sure why I expected a classic bell curve instead of a long right tail.
Edit: playing with the data, looking at some of the sub-distributions, (e.g. front-end engineers with 3-5 years experience at all company sizes) do look a bit closer to normal distributions, but in many, there's still that long tail to the right.
Now, what to do with this information? Perhaps I can play it to my advantage when attempting to negotiate a raise or a salary with a new company.
I was pretty firm on insisting on a bigger raise based on that data, after which my boss's boss came in with the big guns. They had a spreadsheet of compensation data collected by investors in the startup, and this data was collated across all their (and other) portfolio companies.
It was broken down by total funding, company size, years of experience, role, all that. IIRC, it was Option Impact: https://www.advanced-hr.com/. I wish I could've looked at it more, such interesting information with specific numbers. Of course, that will never be made public, much to the detriment of the workers in tech.
They don't criminally underpay or anything, but they are on the low side of average and I've been insisting that they are 20-30% under market on just base salary alone for high caliber people. No RSUs and there's some cash bonuses (maybe up to ~10%) and other near-cash perks but nothing super significant.
Well, jokes on them, I just accepted a new non-FAANG gig for a nearly-25% base salary increase, plus a significant initial RSU grant of ~1.5x salary, plus other bonus potential. And I'm more excited about the company and work anyway.
It's bananas out there right now, go get it while you can.
But I wonder if there are recent studies on the demographics of who attends elite universities in general? I'd assume it is mostly kids from the upper middle class to upper class backgrounds.
These sort of places pat themselves on the back for saying they do target minorities and the less privileged, but they have such an insurmountable background to overcome they may never make it to such schools, despite being smart enough to attend.
You may miss people like: https://www.uidaho.edu/engr/news/features/tom-mueller
While I wont deny the foundational knowledge that CS degrees can offer, it doesn't seem to make much a difference in earning potential.
The biggest difference between CS degree holders and bootcampers / no degree holders is the amount of time + effort required to land that first job. It seems that once you land that first job, there is not a huge difference in earnings between traditional and non-traditional engineers.
Will be interesting to see how it plays out over time. I'm curious what the average earnings will be between bootcampers vs 'others' in 15-20 years.
For exmaple, Amazon has a cap on base (I think it's 160k) and provides huge amounts of stock beyond that. I'd imagine it's a similar story for many companies.
Total comp is a far more realistic and accurate measurement.
These base estimates are much higher than what sites like Levels.fyi and anecdotal numbers I've collected from my new grad peers.
At the new grad level, Base salary numbers are $125-135k:Google, $105-115k:MSFT,Amazon,FB. The base salary numbers at other unicorns don't sound much different either.
I am not saying that TB's numbers are wrong. But, there has to be some reason for the huge difference in the numbers I've heard vs the ones TB reports.
Are you going to refuse a job offer because it pays less than some arbitrary bar, even if it's the best offer you have?
So you see 95% of those salaries are just MS/AMZ
You really think that, with all that, Amazon and Microsoft make up 95% of high salaried jobs in Seattle?
Now, that’s for startups. Banks will pay twice that.
I’m wondering what the medians look like - the means being displayed may be distorted by the finance and Google outliers.
Edited to add: $60k is a middle class salary for a single person in NYC. (Though people who are unaccustomed to small apartment living / the idea of singles having roommates / the idea that middle class means having to compromise on some expenses will complain otherwise.) I wouldn’t want to try to raise a family of 4 here on that income, though.
$130k is definitely upper middle class for a single person, but a family of 4 on that income is going to have to live way outside the hip areas of the outer boroughs in order to be able to afford housing.
The salaries just aren’t that impressive for the cities in question. I can make that in much lower cost of living areas of the country.
I don't know of many engineers with salaries of 200k, but many who will make more than that when you include stocks and bonuses.
I got what I thought was a pretty good low CoL offer as a new grad for $95k base (+maybe $10k-20k bonus?). Haven't heard of people getting such high offers as yours in low CoL areas except for small offices of tech giants
I trust Triplebyte's data to be more representative because it's not one internet identity at a time stating their own number.
Given that the median in SF is $150K, $200k base salary for an experienced engineer is in line with expectations
It seems like only comparing base makes it harder to compare smaller companies (which usually give you all your comp as base) with larger companies (where 30% seems to be rsu and bonus)
