I have zero experience in hiring people; I only know how things seem from the opposite side (as a software developer). The MIRI job announcement seems optimized to scare people like me away. It involves a huge sunk cost in time—so you only take it if you are 100% sure that you want the job and you don’t mind the (quite real) possibility of paying the cost and being rejected at the end of the process anyway. If that is how MIRI wants to filter their applicant pool, great. Otherwise, they seem utterly incompetent in this aspect.
For starters, you might look at Stack Exchange and check how potential employees feel about being given a “homework” at an interview. Your process is worse than that. Because, if you instead invited hundred software developers on a job interview, gave them extensive homework, and then every single one of them ghosted you… at least you could get a suspicion that you were doing something wrong, and by experimenting you could find out what exactly it was. If you instead announce the homework up front, sure it is a honest thing to do, but then no one ever calls you, and you think “gee, no one is interested in working on AI alignment, I wonder why”.
How to do it instead?
Here, my confidence is much lower, but it seems to me that you basically need to choose your strategy: do you have specific experts in mind, or are you just hoping to discover some talent currently unknown to you?
If you have specific experts in mind, you need to approach them actively. Not just post an announcement on your website, hope that they notice it, hope that they realize that you meant them, hope that they decide to jump through all the hoops, so that at the end you can tell them: “indeed, I had you in mind; I am glad you got the hints right—now we can discuss salary and work conditions”.
If you are looking for an unknown talent, you probably want some math and/or programming skills, so why not start with some mathematical or programming test, which would filter out the people without the skills. First round, publish the problems on LessWrong and let people type their answers into Google Forms. Yes, cheating is possible. That is why the second round will be done on Zoom, but only with those who passed the first round. Do not be too harsh at grading, just filter out the people who obviously do not have the talent. What is left is the people who have some talent in general, and are kinda interested in the job. Now you can start giving them more specific tasks and see how well they do. Maybe reward them by internet points first, so they can see how they compare to the competition. When the tasks get more difficult, start paying them real money, so that when it becomes too much of a time sink, it also becomes a side income. (Don’t: “If you do this successfully, I will give you X money.” Do: “Try your best, and I will give you X money for trying, as long as you give me something. Of course if the results are not up to our standards, we may at some moment decide to remove you from the list of participants.”) At the end, either everyone fails, or you will have found your person. Even if everyone fails, you still get some data, for example at which step most of them failed.
Yes, there will be some wasted time and effort. It happens to all companies. Whenever you invite someone for an interview, and after an hour you decide that you don’t want them for whatever reason, you just wasted one hour of salary of everyone who participated in that interview. That is a cost of hiring people.
Importantly: you need to have one person whose task is to do the recruitment. If you just expect it to happen, but it’s actually no one’s high priority, then… most likely it won’t.
I recently went through the job search process as a software engineer who’s had some technical posts approved on the Alignment Forum (really only one core insight, but I thought it was valuable). The process is so much better for standard web development jobs, you genuinely cannot possibly imagine, and I was solving a problem MIRI explicitly said they were interested in, in my Alignment Forum posts. It took (no joke) months to get a response from anyone at MIRI and that response ended up being a single dismissive sentence. It took less than a month from my first sending in a Software Engineer job application to a normal company to having a job paying [redacted generous offer].
I have zero experience in hiring people; I only know how things seem from the opposite side (as a software developer). The MIRI job announcement seems optimized to scare people like me away. It involves a huge sunk cost in time—so you only take it if you are 100% sure that you want the job and you don’t mind the (quite real) possibility of paying the cost and being rejected at the end of the process anyway. If that is how MIRI wants to filter their applicant pool, great. Otherwise, they seem utterly incompetent in this aspect.
For starters, you might look at Stack Exchange and check how potential employees feel about being given a “homework” at an interview. Your process is worse than that. Because, if you instead invited hundred software developers on a job interview, gave them extensive homework, and then every single one of them ghosted you… at least you could get a suspicion that you were doing something wrong, and by experimenting you could find out what exactly it was. If you instead announce the homework up front, sure it is a honest thing to do, but then no one ever calls you, and you think “gee, no one is interested in working on AI alignment, I wonder why”.
How to do it instead?
Here, my confidence is much lower, but it seems to me that you basically need to choose your strategy: do you have specific experts in mind, or are you just hoping to discover some talent currently unknown to you?
If you have specific experts in mind, you need to approach them actively. Not just post an announcement on your website, hope that they notice it, hope that they realize that you meant them, hope that they decide to jump through all the hoops, so that at the end you can tell them: “indeed, I had you in mind; I am glad you got the hints right—now we can discuss salary and work conditions”.
If you are looking for an unknown talent, you probably want some math and/or programming skills, so why not start with some mathematical or programming test, which would filter out the people without the skills. First round, publish the problems on LessWrong and let people type their answers into Google Forms. Yes, cheating is possible. That is why the second round will be done on Zoom, but only with those who passed the first round. Do not be too harsh at grading, just filter out the people who obviously do not have the talent. What is left is the people who have some talent in general, and are kinda interested in the job. Now you can start giving them more specific tasks and see how well they do. Maybe reward them by internet points first, so they can see how they compare to the competition. When the tasks get more difficult, start paying them real money, so that when it becomes too much of a time sink, it also becomes a side income. (Don’t: “If you do this successfully, I will give you X money.” Do: “Try your best, and I will give you X money for trying, as long as you give me something. Of course if the results are not up to our standards, we may at some moment decide to remove you from the list of participants.”) At the end, either everyone fails, or you will have found your person. Even if everyone fails, you still get some data, for example at which step most of them failed.
Yes, there will be some wasted time and effort. It happens to all companies. Whenever you invite someone for an interview, and after an hour you decide that you don’t want them for whatever reason, you just wasted one hour of salary of everyone who participated in that interview. That is a cost of hiring people.
Importantly: you need to have one person whose task is to do the recruitment. If you just expect it to happen, but it’s actually no one’s high priority, then… most likely it won’t.
I recently went through the job search process as a software engineer who’s had some technical posts approved on the Alignment Forum (really only one core insight, but I thought it was valuable). The process is so much better for standard web development jobs, you genuinely cannot possibly imagine, and I was solving a problem MIRI explicitly said they were interested in, in my Alignment Forum posts. It took (no joke) months to get a response from anyone at MIRI and that response ended up being a single dismissive sentence. It took less than a month from my first sending in a Software Engineer job application to a normal company to having a job paying [redacted generous offer].