> Encountered numerous situations where work was "urgent" and would likely land a contract or sales for the company, and everyone would be a superstar if they delivered this "crunch".

> After 2 months of pulling all-nighters and sleeping for 3/4 hours, we deliver the project ahead of time

In my career, none of these have ever paid off. Every time I've crunched this way on something dramatically urgent like this, it has turned out that the "if we can deliver this, this huge moonshot sale is a sure thing" turns into a no-sale

The sales person never seems to get cut loose for diverting the entire R&D towards a longshot for months and burning people out, though

And you can bet the sales person isn't putting in weeks of overtime for the duration, either

I basically refuse to do overtime anymore unless I'm working extra to make up for my own screw up. I'm not putting in extra to hit some other assholes unrealistic deadlines ever again

Agreed. Even if by some miracle you do deliver, and are considered a superstar, then what? What do superstars get? Probably just even more crunch work, since you've proven you're willing to do it.

You imagine that it will catapult you ahead in your career, your income will skyrocket, you will be respected and loved by your company and peers

But in reality no one really cares much, you'll get the same raise everyone else gets, your bonus is still gonna be capped by your contract, and you will be better off finding a new job if you want more money

Man sometimes I want out of tech so badly it hurts but I don't generally think it's better anywhere else

This reminds me of the scene in Schindler's List where the SS officer asks the enslaved factory worker to show him how fast he can assemble a particular component. The terrified worker races to assemble it in record time, anxious to please and impress the nazi -- who responds to the effect of: "if you can make them that fast, why is your daily quota so low?"

This was also a standard technique on American plantations, then adapted to the industrial economy in the form of Taylorist time-and-motion studies. If you trace modern management practice, it is basically a straight line back to chattel slavery.

The plantation thing was so horrible. They would track your personal best output, and every day you didn’t beat it you would get whipped based on how much you fell short. Of course this made you work faster, but it was like a game of 21 because you knew if you went over you would now have a higher quota to meet from then on.

Absolutely 0 had ever paid off. Probably worst was trusting too much a colleague perceived by everybody as Oracle/plsql guru, when troubleshooting vendor's abysmal performance of DB queries during some bigger migration (up to half an hour easily, for trivial 30 million rows). He didn't see any issue on DB side, pointed to useless oracle hints, crappy JDBC drivers, spring's jdbc templates, possibly my not-optimal code etc.

I went over my head, did probably the most complex code in my life, massively parallel, over weekends and evenings. That wonderful cathedral didn't move performance a zilch, just made debugging and further changes much harder. After few hours of actual debugging afterwards he found out vendor defined responsible DB table in such an obscure and bad way way that we had to literally copy whole table to another more sane one, and perform all the work there in maybe 5% of the time. In fact I suggested exactly same thing initially but it was quickly dismissed by him, and who questions the guru, right.

This didn't even come from management just colleague's incompetence/ego, hard deadlines, tons of pressure to deliver, and starting project already 2 months late. Closest I've been to burnout yet. I am still a bit pissed off on him, but I know it was not malice so that eases emotions quite a bit.

And to similar request coming from the top - been there, done that too, regretted that time & energy put in it. These days, 8 hours days, if I am not making it on time, I communicate early & clearly and that's it. They handle it, and if they don't, well there is always next job. Life is about priorities.

This is the hardest lesson to learn. Sometimes you won't be afforded the ability to do it "right", either for the company or the product or the customer. Eventually, you'll decide to just show up and ask what is most important today and work on that. Then clock out completely when your work is done and go find meaning and personal satisfaction in your personal life. Go exercise or volunteer or get a hobby or be present for your family. The best way to have work/life balance is to separate them. This is also one of the reasons why I hate wfh. The drive to/from is a great separator and decompressor for me.

I think this is so hard to learn because it's counter to human nature, and only necessary due to the artificial conditions of the modern world. We're programmed to want to be useful to our tribe. But we don't live in tribes any more. Our brains get confused and burnt out because we perform and perform and perform, but we don't get love and status and security in return, we just get this abstract thing called money, which it doesn't really understand.

We just need to put ourself first and our health