[humanrebar]

> Learning a new skill is not an earth-shattering problem...

Depends on the problem. If you're cranking out brochure websites from a set of templates, you have a pretty accessible role to fill.

But I have many colleagues who make mistakes (even when people object clearly), dig a hole for themselves, double down on the mistake, and the either back their way into job security with their mess or end up moving on once the chickens come home to roost. Some of the time, I can even see their eyes glaze over as I try to get them to think a few more moves ahead in their plan.

Now, organizations should have safe roles for journeyman engineers who crank out and maintain code as directed, but most organizations expect all senior engineers to be part time system design experts, which I don't think everyone has the knack for. Likewise, not everyone is a great communicator and strong enough technically to be accurate in their communication.

[crispyambulance]

    > But I have many colleagues who make mistakes [...] dig a hole for themselves, double down on the mistake, and the either back their way into job security with their mess or end up moving on once the chickens come home to roost.

A skilled practice of behavioral interviewing can, I think, mostly screen out such candidates. People who have a habit of "doubling down on their mistakes" will find it hard to wiggle around pointed behavioral questions which require the candidate to describe real experiences and answer follow-up questions.

The problem is that most organizations are terrible, TERRIBLE at evaluating candidates. They forget that interviewing candidates is a skill in itself that needs to be developed and needs to be consistent across the organization.

Moreover, these same organizations fail to "close the feedback loop" when it comes to evaluating employee performance. The "annual review" data can be put to good use by informing hiring decisions. Instead annual reviews are just an irritating burden used for the purpose of carving up the raise/bonus pie. It is basic engineering... evaluate your outputs (employees), then use that data to adapt to your inputs (candidates). But virtually no one does it, because its all tied up in HR-bullshit instead.

[BeetleB]

>But I have many colleagues who make mistakes (even when people object clearly), dig a hole for themselves, double down on the mistake, and the either back their way into job security with their mess or end up moving on once the chickens come home to roost.

Your making the parent's point for him. This is a behavioral problem. Hence, screening for culture fit.

[humanrebar]

> This is a behavioral problem

No. These people are generally open to suggestions that they grok. They just can't, at least sometimes, comprehend more indirect or abstract problems (like CAP theorem implications that influence the design of SQL tables and business keys). They write code that meets requirements in the short run but get bogged down in a local maximum that can't be improved upon without an expensive rewrite.