I think it makes more sense to hire people you like who you think can do the job. Then train those individuals to be the employees you want. That is where your diversity will come in. What happened to businesses and people investing in and training other people? Sure they may walk in a dud in your eyes but there is no reason why they should stay that way.
Secondly, many companies can take a good engineer, not use them correctly, run them off or turn them into a bad engineer for your company. There is too much emphasis on what a candidate already is and not enough emphasis on what you can mold your candidate to be after they are hired.
Training people on the job is where your diversity will come from. I noticed that many people got in the door with less skill than the people they are hiring.
From a purely utility-maximizing perspective, it's easy to see what happened. Training people costs time and money, both in non-productivity of the new person, and lost productivity from the people training them. Also, there's a fair to good chance that once you train a new person and they get a couple years of experience, they will jump ship and leave you in the same position as when you started. If you look at it this way, it's better to hire a person who can already do the job than it is to hire someone who might be able to do it with a lot of training. (And yes I know it's a chicken and egg problem of who betrayed whom first with respect to long-term employee retention over the last few decades).
The notable exception to this thinking is in established firms where there is a specific methodology and persona they wish all employees to have. Ex. the major consultancies, law firms, investment banks, etc. For them, it's generally best to pick a completely fresh person (new grad) and mold them into exactly what the company wants without having to "undo" things they may have learned elsewhere.
I can certainly see where you are coming from. As you stated, it is a chicken and egg problem somewhat.
However, if the person has no background whatsoever that's actually different than someone who has some verifiable experience but just isn't what you would like initially.
Wouldn't it make sense to take in someone who has some experience then train them. Sure they may leave but they may not if you treat them well. An employee leaving in the tech world is always a risk. I notice many companies spend all of their efforts hiring but very little effort is exerted to retain people.
At the end of the day the tech world could stand to give back to the community in some way. If it means taking on a bit more hiring risk, its a small service to the communities they've taken so much from.
Completely agree with you - there is a happy medium with respect to experience and training that the vast majority of hiring completely misses out on today. I think the problem is that's it easier for hiring managers to keep a position open for longer while telling themselves they need a unicorn than it is for them to actually put in the effort to train a new person properly.
And yes, employee retention is a constantly debated topic that most companies only pay lip service to.
Secondly, many companies can take a good engineer, not use them correctly, run them off or turn them into a bad engineer for your company. There is too much emphasis on what a candidate already is and not enough emphasis on what you can mold your candidate to be after they are hired.
Training people on the job is where your diversity will come from. I noticed that many people got in the door with less skill than the people they are hiring.