|
|
|
|
|
by aristofun
734 days ago
|
|
I prefer reasonably scoped take home assignment to leetcode "prove me you're awesome in 30 minutes" kind of interviews. Relying on these kind of signals as a main input is a sign that company didn't think through their hiring process and rather just cargo-culting. But all that are still secondary concerns: good people can make bad process work, bad people tend to ruin the best of processes. So for me the only important "flags" are specific people I contact — their attitudes, levels of professionalism, questions they ask, answers they give, their decision process etc. On all steps of the interview process. |
|
Some people I've worked with in the past scoff at this idea. When I ask them how they would prefer I evaluate technical ability they seem to all meander around the question basically saying you know a good hire when you see one.
It's tough to hire good people. I've been in the hiring process for Stanford graduates who could barely code in their take-home. I've also hired candidates who freshly changed their careers later in life and wowed us with their take-home. The 4 hour take-home for us has been wildly helpful in identifying the best hires, but I do get it can feel like a burden to some.
I just don't think I could really trust the outcome of leetcode. To me, I feel like anyone can grind leetcode for a couple months and ace a test on it. Does that mean they can create anything of value? Not really.