3 ms·

This ignores something very important: tribal knowledge. Truth is, losing a tenured engineer carries a MUCH larger cost than HR and some managers seem to under

This ignores something very important: tribal knowledge. Truth is, losing a tenured engineer carries a MUCH larger cost than HR and some managers seem to understand-- these engineers understand the nuances behind why certain decisions were made, what kinds of things tend to work and don't, minutiae of various systems that simply will never get documented, etc. Not to mention the influence they can carry in the org. You're not wrong that a lot of engineers stagnate after a few years... but it's a big logical jump to put that blame on them rather than the org they're joining. That said, I like to always call out that there are "developers" and there are "engineers". The majority of software roles call for the former and these tend to be more "replaceable". It's harder to say that for the latter where they are involved in more strategic technical decisions.

randcraw 5y ago
The value of tribal knowledge is even more apparent when 'engineers' describes a) traditional engineers (EE, mechE, chemE, civilE, etc), or b) software types with deep specialized skills that grow with domain experience (e.g. image analysis in medical/pharmaceutical, signal processing in audio/radar/telecom, or finite element analysts in weather/mixture modeling). The experience gained with years in roles like these isn't just in the use of computer tools, but in how to approach and solve proprietary domain problems with constraints that often aren't obvious. Knowing the pitfalls or conventions inherent in complex or regulated domains like automobile or pharmaceutical safety takes time to learn, making those with such experience more valuable to retain.
rightbyte 5y ago
> a lot of engineers stagnate after a few years Stagnate is also a way of saying got good at the job. I mean after a couple of years at a place an engineer might be the person that know the systems best in the whole world if they are no commodity. It takes year to gain the experience and inside out knowledge of a complex system. To me it feels like I get a mental picture over a system after some time where I can smell where bugs comes from and very fast decide were functionality should be inserted. Up to that point things are just a hassle and then it clicks and becomes a breeze. Then I switch jobs to get a pay raise and the grind begins again ...

randcraw 5y ago

The value of tribal knowledge is even more apparent when 'engineers' describes a) traditional engineers (EE, mechE, chemE, civilE, etc), or b) software types with deep specialized skills that grow with domain experience (e.g. image analysis in medical/pharmaceutical, signal processing in audio/radar/telecom, or finite element analysts in weather/mixture modeling). The experience gained with years in roles like these isn't just in the use of computer tools, but in how to approach and solve proprietary domain problems with constraints that often aren't obvious. Knowing the pitfalls or conventions inherent in complex or regulated domains like automobile or pharmaceutical safety takes time to learn, making those with such experience more valuable to retain.

rightbyte 5y ago

> a lot of engineers stagnate after a few years Stagnate is also a way of saying got good at the job. I mean after a couple of years at a place an engineer might be the person that know the systems best in the whole world if they are no commodity. It takes year to gain the experience and inside out knowledge of a complex system. To me it feels like I get a mental picture over a system after some time where I can smell where bugs comes from and very fast decide were functionality should be inserted. Up to that point things are just a hassle and then it clicks and becomes a breeze. Then I switch jobs to get a pay raise and the grind begins again ...