mm, perhaps but I'm echoing documented hiring philosophies of the companies. They are aware they make mistakes. But they feel confident they filter out a majority of false positives along with many false negatives. They have their own metrics for determining this in relation to how many people they hire are able to complete a minimum satisfactory work.

They have documented people who can do this in relation to the people they hire, and even have internal portfolios tracking employees by age, experience, major, and schools, as well as how they performed on interviews. This is a fact.

Everything is subjective. But ultimately a company decides based on its own subjective priorities how an employee meets up to their subjectively defined needs for the position that person fills, and they can pretty onjectively within that context evaluate whether the person can do the work or not, and even track commitments, errors, time to cdoe to completion, amount of bugs, amount of time team has to spend fixing their bugs, etc, and parse out a pretty good idea of performance.

Thats why theres multiple books written by these companies designed to help you pass their interviews, as they have identified indicators of low performance in interviews....

All companies can do this, but many don't, and many arent based in software and therefore do not have embedded ability to track performance metrics the way code commits can.

small companies can do it too, but big companies who also have this system in place can take advantage of larger sets of data for more accurate indicators that seem to correlate with low performance.