It sounds like it's either you take part in the life-consuming startup-founder lottery or you go work in BigCo (possibly not even tech) for a good life until you see a worthy opportunity. Or you start doing freelancing/contract jobs till you have a steady stream of income, which is sort of in-between. Anyone have advice?
No, author's case is severely overstated.
In my experience startup jobs in SF/SV/NYC pay as much as, if not more than most big tech company jobs. The notable exceptions are Google and Facebook, where the compensation and stock is extremely good and generally ahead of the rest of the pack.
Which isn't to say that author's complaints about crappy startup jobs are false, but rather that they're really not unique to startups at all - they're common across all tech jobs, regardless of company size.
In reality there are three classes of tech jobs:
1 - Near-perfection. Great opportunities for advancement, great pay, great benefits, great work/life balance.
2 - Somewhat a pack of lies. You'll either work too much, get paid too little, get to fight management all day, have little room to grow, or some combination thereof.
3 - Complete pack of lies. Deranged management, just complete rubbish.
Most tech companies, even big ones, fall into the second and third categories. I think author is extrapolating a bit from the presence of Google in his seeming belief that most big tech companies are great places to work. They are not.
Case in point:
> "You can learn more by being in a company that offers reasonable hours while you can have a side project. An open source project, a small app, something your parents can use, anything that you either enjoy, try to make it big, or just make the world a better place."
Most big companies aren't like this. In fact, I can count the number of such companies I know on a single hand and have a finger or two left over. Most startups aren't either, so more accurately most software jobs aren't like this.
The blog post seems to be a lament about the shitty quality of startup jobs, but really it's more accurately a lament about the shitty quality of software jobs in general.
He wants startups to have good compensation, good environments, and reasonable hours - but most big companies can't pull this off either. Big Corporate software jobs are infamous for low compensation, infamous for being middle-management-infested hellholes, and just as willing to bend hacker culture to its own ends.
(Thanks also to the other kind commenters who replied - ctide, rayiner, smartician, bkanber, prostoalex, avenger123, and akurilin - your answers are valued!)
Google has a lot of great engineers, and if you get a good manager for your first project and get to a launch in your first 2 years, you become a Real Googler and it's awesome. If you don't establish credibility early on, you're fucked. It's impossible to transfer to a good project without already having had a good project, so your first manager determines your career.
On the whole, I think Google is in decline and it has an absolutely shitty management structure (probe about "calibration scores") but there is still some genuinely interesting work and there are some great engineers so, by all means, give it a try.
However, there's probably a 60% chance that you'll land on a dogshit project with no career advancement and have to job-hop yet again in a few months. Make it 85% if you're outside of MTV. But if you don't mind taking that risk, then try it out. Google's upside (stable pay and great projects) when you actually land well is too high to write it off.
People who works at normal places (and have worked at normal places) don't have any inclination to write blog posts about how not awful their employment is, only people who have bad experiences.
When interviewing, I'm worried about a)Seeming like a "slacker" for inquiring about expected hours. I want to convey that I don't mind working late for an urgent issue or close to release time, but that I can't sustain >50 hours b)how do I know they're being honest with me about working conditions?
Places with sane hours think it's a reasonable question.
Spikes to 50-60 happen and, if you're in a position of ownership, you better be prepared to throw down when they do. But no company should ever set the baseline at 50 hours per week. That's counterproductive and it leads to burnout and catastrophe when spikes occur.
I actually believe in planning based on a 3-hour work day. A 3-hour block of uninterrupted time to focus on work is an expectation and also an entitlement. The other 5? Spend those catching up, learning about other projects, and building skills.
1) Compensation was very solid (this was pre-Bubble 2.0 in a very low cost of living city); 2) We got full benefits (zero employee contribution for either health insurance or dental); 3) We got off the entire time between Christmas and New Years independent of our vacation time; 4) Most people worked ~50 hours/week, except before a new release or an important demo (but this wasn't all the time)
The company never took VC money, wasn't looking to exit, but generated a good living for the founder (going on 13 years now).
If you're looking for a solid job without killing yourself, places like that exist. They usually work in specialized domains with stable sets of clients.
I'm at one of those companies, and there are times where I feel a bit like a second rate developer, but I work less than 40 hours a week, and I still get my six figure salary, plus 20+ days of vacation that I can actually take.
Size isn't the big issue. The reason VC-istan jobs tend to be so rotten is that the companies are red-ocean marketing experiments and technology is a sideshow. It's hard to grow at the 200%/year rate that "ooh, shiny" VCs expect without shitting all over your culture. Mid-growth (10 to 40% per year) is the future.
What's the best way to find those 20-100 person mid-growth startups? I'm looking for machine learning and functional programming, but I'm old enough (29) that I don't give a third of a rat's ass about "prestige" or "hipness".
Here's my observation: companies tend either to be r-strategic or K-selective (in the evolutionary biology sense). The mid-growth, quality-oriented companies are K-strategists. The get-big-or-die gambits and the MegaCorps are r-stragetists. Now, the major problem in marketing yourself to the mid-growth K-strategists is that they hire slowly, meaning that you need to be an exact match for a role they need right now. There isn't the aspirational "draw in talent and figure out what to do with it later" hiring of r-strategists. I don't think the mid-growth economy is fleshed out enough to provide the job security of VC-istan disposable-company job-hopping.
As a non-resident you are pretty much at the mercy of whoever is sponsoring your H1B. If you are from the lucky set of countries and have a graduate degree, you can sit in purgatory at a BigCo for 5 years and then finally have the right to quit and start your own gig.
The H1B is a tough obstacle, and I'm instead looking at the 18-month renewable H1B1 which applies to workers from Singapore and Chile and doesn't lead to permanent residency. I'll have to look for people who've made the jump and get their advice, I guess. And network with people in good companies which might be hiring.
We have a "no overtime" policy at my company (unless you really want to), and everyone's paid and treated well. We value emotional quality and mental health, and would rather stick to 40 hour weeks than have people burn out. And there are lots of others like us!
Edit: however we're neither in the valley nor have we taken VC funds.
1) Large companies typically dedicate more attention to code quality and review process to ensure all of the low-hanging gotchas are covered.
2) Larger companies, if not adherents of test-driven development, typically still rely on large suites of unit and integration tests, and continuous deployment, that follows from having all the tests in order.
3) Larger companies typically have processes to analyze failures in dev/staging/production, and resources to throw at mitigating those failures in the future.
With all that said, there's nothing to prevent small companies from instituting such practices as well, and it's typically the work of the first few senior engineers the startup hires.
When you interview with the companies, remember that interview is a 2-way street - you interview them as much as they interview you. After you're done with obligatory fizzbuzz and binary search questions, ask them about the deploy processes, code review processes, latest fires, and other reasons someone would be in the office at 1 am at night (other than by their own choice).
In terms of all startups being like this, its a big "it depends".
There are many companies doing software development that are across the full range of the spectrum. Don't be assuming that the BigCo or the startup represent the full range.
Since it looks like you can be somewhat free in deciding where you live, you have a big advantage. Making a good income doing software development means different quality of life depending where you are in the US. Perhaps your dream of moving to SV might be to overreaching. So break it into steps. Focus on getting some kind of software work in the US in a good place. Once you are here, you might just change your mind about what you want to do.
She probably has a Green Card or Citizenship by now, but if I recall correctly she was on an H1 visa at the time. If she lost her job for any reason, she would have had to find another employer pronto or she would have had to leave the country.
Her employer did not find it hard to ask her to work from 8am to midnight 7 days a week.