[UncleMeat]

I think a key error people often make is assuming that the line engineers are always right. If a decision pisses off the line engineers it must be idiotic! The manager must be a fool. Peter principle in action.

This surely happens sometimes. But I’ve personally witnessed a lot of engineers (especially junior engineers) who don’t give a shit about customers or shipping products or building alignment or communicating planning and impact. And to have a successful team/product/company you will need to aggravate them sometimes and they’ll post things on medium about how their boss sucks and that nobody should ever estimate anything or how all tech debt is awful or whatever.

Good managers should support their team’s growth and careers, which should hopefully eliminate some of the “my manager is an idiot” stuff but it can’t eliminate all of it.

[amznthrowaway5]

> This surely happens sometimes. But I’ve personally witnessed a lot of engineers (especially junior engineers) who don’t give a shit about customers or shipping products or building alignment or communicating planning and impact.

This attitude is what I've usually seen from management, and junior engineers are often the ones who fight against this attitude, before they get too jaded and leave. Management will dishonestly claim successful delivery with zero regard for the actual product, and then force the engineers into 24/7 oncall.

> Good managers should support their team’s growth and careers, which should hopefully eliminate some of the “my manager is an idiot” stuff but it can’t eliminate all of it.

That's great, but it's no substitute for actual understanding of the problems. Planning done by incompetent managers is nonsense since they don't understand the low level details of the problems. From what I've seen, promotions into management are usually related to being best buddies with others in upper management. Incompetent managers will then use all types of excuses to justify their position, "I'm focused on hiring and growth of the team", "It's hard to be responsible for people", "leadership skills matter!" etc. all the while hiring and promoting their fellow incompetent friends and having zero care for the actual product.

At least that's what I've seen at Amazon across multiple teams. Over here the incompetent management cliques will even hunt for the junior engineers brave enough to give the managers bad scores on the daily "connections" pop-up survey.

[majormajor]

> This attitude is what I've usually seen from management, and junior engineers are often the ones who fight against this attitude, before they get too jaded and leave. Management will dishonestly claim successful delivery with zero regard for the actual product, and then force the engineers into 24/7 oncall.

Or do junior engineers build a broken system because they don't know better which results in them having to constantly fix shit and burn out?

And then a senior engineer is hired from outside with a lovely mess to clean up while getting paid more than the junior engineer was. And at that point in the company's life, the pressure is usually lower. So it's a nice gig. ;)

[amznthrowaway5]

That’s possible but not what I’ve seen. I’m on ML teams, the junior engineers often come in with a good CS background and sometimes good software engineering practices, while the highly credentialed senior scientists often refuse to even write unit tests.

[HomeDeLaPot]

That's ML. Senior scientist != senior engineer, right?

[amznthrowaway5]

Well it's Senior "Applied Scientist". They get to make all the rules and get a much higher salary than sr engineers. ML teams seem to consistently reward the wrong people who can't actually deliver real value.

[smallcharleston]

They might or might not deliver real value, but I’m sure most research scientists view a lot of software as simple implementation rather than real value. I think everyone needs to work together for these complex systems.

[UncleMeat]

> Management will dishonestly claim successful delivery with zero regard for the actual product, and then force the engineers into 24/7 oncall.

There is a balance here. This happens I'm sure. But I've also seen healthy velocity (no mountains of debt, no oncall) be pulled down by engineers who insist on working on what they find fun rather than what is actually good for a product. And these engineers have called their managers idiots to me.

[sidlls]

I’ve seen what you describe in engineers at all levels, mostly evenly distributed. Engineers just tend to be very narrowly focused on technical solutions. That’s a problem both in management and culture in the industry.