This is very true. We weed out hundreds of applicants ler opening on very simple questions... Think what is the difference between const, var, and let declarations in JavaScript?

The problem is we're all spending weeks and weeks on weeding out fakers. We've moved to a quick pop quiz right after we get an apply. This takes out 85% of the candidates (it makes me sad when I look at how simple the stuff is that weeds them out...). Two interviews are more than enough to land on a hire after that.

Our company is a conversational recruiting software company (you apply, we say hi, and screen candidates in real time and then hook the candidates up with an interview). We started doing the screening in our chat flow and it works really well for us.

Very good advice. The "gatekeeping" they alluded to is just very necessary if they want to have any differentiation from any other generic job matching service. I'm sure the triple byte team is aware of this but I think it's important to keep in mind just how much people work in tech/IT jobs, how wide the skill/qualifications range is and how spammy applications can get very, very quickly. Even positions that are "easy" to get as a competent engineer can get hundreds of applications, and since it's also pretty common for people to just outright lie on their CVs it becomes very time consuming to filter for someone who genuinely meets even just most of the job requirements.

& maybe I'm wrong but I thought the entire value proposition of TB was that their pre screening and testing process made the whole process much less noisy for your hiring managers. If you still end up needing technical interviews and aren't guaranteed that all the candidates you get have gone through some technical screening, what's the point?

AIUI, employers will be told whether the candidate took the tests, and if so what their score was:

> Going forward, our assessments are a purely optional means by which engineers can show their skills to employers (who overwhelmingly tell us that they trust our scoring), rather than a requirement for entry.

I think the idea is that an employer can still filter out the unfizzbuzzables, but has the option to ignore that for candidates whose stellar track record is a stronger signal.

I'm a little bit skeptical, because i think there are lots of people with apparently strong CVs who are still weak when you actually put them to the test. But maybe not at the very top end; i am really thinking about people with experience in various investment banks, which is perhaps less of a signal than Apple then Netflix or something.