Richard Möhn

Karma: 268

Be careful with (outsourcing) hiring

Richard Möhn17 Oct 2022 20:30 UTC

40 points

38 comments13 min readEA link

Richard Möhn 17 Nov 2022 1:55 UTC
36 points
2 ∶ 0
in reply to: Milan_Griffes’s comment on: Who’s at fault for FTX’s wrongdoing
Some corrections of the Sequoia info:
- I’ve never been a grad student.
- I’m neither Japanese nor a Japanese citizen.
- I ‘volunteered’ in the sense that people at Alameda reached out to me, I said ok and then got paid by the hour for my help.
- ‘(obscure, rural)’ is an exaggeration. ‘provincial’ would be a more apt adjective for the location. The main bank we used was SMBC, the second-largest bank in Japan.
- ‘for a fee’ sounds as if it was some sort of bribe to get them to do what we wanted. But we only paid the usual transaction fees and margin that any bank would charge.
But mostly, if https://forum.effectivealtruism.org/posts/xafpj3on76uRDoBja/the-ftx-future-fund-team-has-resigned-1?commentId=hpP8EjEt9zTmWKFRy is accurate, I’m bummed that the money I helped earn was squandered right away.

Richard Möhn 6 Feb 2020 11:36 UTC
23 points
0 ∶ 0
in reply to: Habryka’s comment on: Request for Feedback: Draft of a COI policy for the Long Term Future Fund
Some suggestions:
- Go over this with a lawyer and let them formulate it right.
- Replace ‘romantic and/or sexual’ with ‘intimate’.
- Right now those terms are at the top level of the bullets. One could make them stand out less by turning the hierarchy around. Example:
Sufficient causes for recusal of a fund member: Applicant is a close family member, or is/was a romantic/sexual partner within the last year.
Not sufficient for recusal, but should be made public: Intimate relationship that lasted longer than two weeks and ended more than a year ago. / Current intimate relationship with a third person who is in an intimate relationship with the applicant.
- Pull things together to make them more innocuous:
The fund member has shared particularly intense experiences with the applicant on the level of, eg. a week-long silent meditation retreat, a CFAR workshop, or a shared psychedelic experience.

Richard Möhn 6 Feb 2020 6:13 UTC
21 points
0 ∶ 0
on: Request for Feedback: Draft of a COI policy for the Long Term Future Fund
Who would get to see this policy statement? If I imagine myself in the shoes of a more conservative potential donor, who is checking the fine print as part of their due diligence, I would be put off by phrases like ‘metamour’, ‘consumption of drugs’ and the repeated mentioning of sexual relationships.

Richard Möhn 16 Nov 2022 22:43 UTC
18 points
8 ∶ 1
on: Kelsey Piper’s recent interview of SBF

I have 2 weeks to raise $8b

that’s basically all that matters for the rest of my life

Two weeks to live? I’m slightly worried. But only slightly.

[To clarify: I wrote ‘slightly’ because it appears that Sam writes a lot of things that aren’t very reliable and well thought-out. So I didn’t want to put too much stock in a particular phrasing.]

Richard Möhn 15 Nov 2022 1:53 UTC
15 points
4 ∶ 0
on: How the FTX crash damaged the Altruistic Agency
Please forgive the insensitivity of the following question. I’ve been wondering about this for a while and now I’m (ab)using your post about your troubles and ambitions as a place to ask it: What is your thinking on charging money for the tech support part of your work?

The EA ecosystem appears to consist of organizations supporting each other for free, while being funded by a third party. How about third-parties funding only those organizations that do direct work and can’t have paying customers, and having the rest be semi-commercial, organized by market forces? (I know that this wouldn’t work for your more ambitious projects. But for the day-to-day tech support it should be possible, at least in a market economist’s dream world.)

Richard Möhn 23 Aug 2022 8:04 UTC
11 points
0 ∶ 0
on: Perhaps the highest leverage meta-skill: an EA guide to hiring
<First sentence redacted.>
The above article, too, contains many questionable pieces of advice. Just one example: ‘Craft your questions so that it’s not embarrassing or difficult for candidates to admit that they don’t have the personal qualities that you are looking for.’ So you’re supposed to rely on candidates admitting that they are not right for the job? Take the article’s example of risk-taking – why wouldn’t you write in the job ad that risk-taking is required? Then the honest non-risk-takers can sort themselves out before going through the trouble of applying.
What about those who don’t sort themselves out? According to the article, you ask them a sort of trick question: ‘are you somebody who likes to jump head first into a problem or do you hold more of the philosophy of looking before you leap?’ This can work if you’ve successfully left the candidate in the dark on whether the job requires risk-taking. If the candidate knows that risk-taking is required and really wants the job (and who doesn’t?), guess which of the two alternatives they will pick.
Talking about going through the trouble of applying – the above article doesn’t address the feedback about hiring that made a splash on this forum a while ago: https://forum.effectivealtruism.org/posts/jmbP9rwXncfa32seH/after-one-year-of-applying-for-ea-jobs-it-is-really-really I don’t agree with the attitude of the latter article, but it does contain important points that ‘an EA Guide to Hiring’ would do well to address.
Addendum – a better question to find out about a candidate’s risk-taking. Say: ‘It’s hard to have a big impact without taking risks. Tell me about a time when you had to choose between a safe and a risky course of action. What were your considerations, how did you implement your choice and how did it turn out?’ After they’ve given an answer, probe. Ask follow-up questions. Always about the past. If all you dig up is more details about successful risk-taking in the past, there’s a good chance the candidate fulfills your risk-taking requirement. On the other hand, if they don’t have done much risk-taking, you will find out quickly using this course of questioning. This is not to say that you should turn your desk lamp on them and play a Stalin-era interrogator. All the points about being extremely friendly and putting the candidate at ease still apply.
Note to those who are downvoting this: Please describe your disagreements in a comment. Otherwise nobody can learn from them.

Richard Möhn 21 Jan 2023 9:49 UTC
10 points
0 ∶ 0
on: Available talent after major layoffs at tech giants
Great reminder! This podcast episode has some actionable ideas for how to reach out to people who were laid off: https://www.manager-tools.com/2009/02/bench-development-downturn

Richard Möhn 1 Dec 2022 9:09 UTC
10 points
1 ∶ 0
on: SBF’s recent live interview at the DealBook Summit
According to https://finance.yahoo.com/news/ftx-japan-unit-drafts-plan-124544036.html, FTX Japan has about $150 M in assets, which isn’t much compared to what the whole FTX conglomerate owes.

Richard Möhn 27 Aug 2023 3:28 UTC
9 points
1 ∶ 0
in reply to: Larks’s comment on: Soaking Beans—a cost-effectiveness analysis
Perhaps it’s because of farts? I hope I don’t get downvoted for bringing this up – it’s serious and, I suspect, a major inconvenience for people who cook beans from scratch.

It could be that cooking beans for a long time (maybe combined with regular skimming or whatever other techniques we don’t know about) makes them cause way less flatulence than soaking them in little water and cooking for a short time. People recommend soaking for reducing beans’ gas-generation potential, but in my experience (yes, I’ve used a tally counter to count my farts under different soaking and cooking conditions and bean species, but obviously N=1) it takes a lot of soaking water to do that. I’ve found 7 l for 500 g of dry beans to work decently. After soaking I drain and then cook in a small amount of water.

And soaking with a lot of water has implications that might not work for Ugandans:
- You need a much bigger soaking container than cooking container. (Or have to soak in multiple changes of water, which becomes tedious.)
- You need more water.
- If you want a lot of beans for dinner, draining the water becomes physically difficult for one person.
Speaking of dinner: If you eat three meals with beans per day, you’ll have two or three containers going for soaking. It’s likely easier to have one pot over the fire that just gets restarted soon after each meal. (Assuming that refrigeration is hard to come by in Uganda.)

Further practical issues:
- If the beans are small, draining the soaking water requires a sieve or colander, which people might not have.
- Beans start to ferment when soaked in hot weather (as mentioned in other comments, I think). The more they ferment, the stranger they taste and the longer they need to be cooked because of acid building up. (Try and cook some beans with vinegar added to the water.)
- It’s easy to forget to start soaking them in time.
Personally, I do soak and use a pressure cooker. But perhaps that’s a luxury.

Richard Möhn 18 Oct 2022 9:10 UTC
9 points
0 ∶ 0
in reply to: MichaelA’s comment on: Be careful with (outsourcing) hiring
It would be nice if you moved your last paragraph first. Recommending (in bold) based on a casual listening not to read a post feels unfair. (Speaking about fairness, there are other posts about hiring (eg. https://forum.effectivealtruism.org/posts/XpnJKvr5BKEKcgvdD/perhaps-the-highest-leverage-meta-skill-an-ea-guide-to#comments) that are argued more thinly than my post and have gotten much less criticism.) I agree with your alternative recommendation to read the post with skepticism. That’s the case with any post.

I agree with some of your points. Especially the one about ways to outsource without knowing about the subject. I might work it into the article. (Two sentences added 2022-10-19.) (2022-10-22: I’ve worked it into the article.) Overall I think there are misunderstandings, which I should have worked harder to avoid.
- The main point is not hiring advice. (2022-10-20: Added clarifying comments to this end.) The main point is that people who hire need to learn more about hiring. I’m merely demonstrating this by countering Hirely, which I do based on hiring guidance that I’m fairly confident is better than theirs. How this came about I describe in the section about the article’s history.
- I agree that it would be nice to have more epistemic nuance and diligence. But this article wouldn’t exist had I tried for that, because I don’t have more time. Given the choice between not publishing the article and publishing it with some epistemic hubris, I thought the latter is more beneficial.
- The main resource for this article is Manager Tools because I think their hiring guidance is an excellent starting point. This is based on wider reading and listening and thinking and experiencing, which I have been doing for years. I agree that their arguments are often sloppy and there are areas other than hiring in which they’re weaker than they admit. Your claim that ‘they also usually provide no evidence or arguments for their claims except their personal experience’ is wrong as it stands. They provide evidence for some of their claims, which I’ve linked in my section on why Manager Tools is a decent authority. (Note that I wrote ‘a decent’, not ‘the ultimate’ – epistemic humility.) I wish they would provide evidence for other claims, too. And the bulk of each podcast/whitepaper is made up of arguments for their claims, spotty though they might be at times. Personal experience can be a valid support for arguments, too, if it’s believable that you have a lot of it. It is believable that MT has a lot of personal experience (and data) about hiring, which I also point out in my section about them.
By the way:
- I do think I’m fairly transparent and indicate levels of confidence to some degree, perhaps in a different way than you’re used to. I don’t have time to read and digest the full Reasoning Transparency article. But if I look at the top recommendations:
  - ‘Open with a linked summary of key takeaways.’ – That’s the title in my article.
  - ‘Throughout a document, indicate which considerations are most important to your key takeaways.’ – With some squinting that’s the table of contents in my article.
  - ‘Throughout a document, indicate how confident you are in major claims, and what support you have for them.’ – I do indicate my support (Manager Tools) and why I think it’s a support. I am quite confident in my main claim and in the main supporting claims. Note also that I don’t write that outsourcing hiring is terrible, but that you need to be careful. And I don’t write ‘Hirely will cause you to make a bad hire’, but that they put you at risk of making a bad hire. Among others.
- I think I’ve refuted part of your claim about personality tests in other comments. Added 2022-10-22: Even if Big Five assessments can predict some things well, I would expect that to only be the case when they’re not gamed. The one I know can easily be gamed. (Eg. just answer ‘very accurate’/‘very inaccurate’ to all the question that sound like pushing up/down the conscientiousness score.) And an important part of hiring is keeping out people that make themselves look better than they are.
- Expanded 2022-10-23: Some of the arguments against the ‘bad’ hiring advice had already been addressed in my article. Eg.: I point out that the résumé screen is a flawed filter and needs to be supplemented or replaced, depending on the situation. I briefly argue why it’s okay to take a lot of time interviewing. There is also an easy fix for anchoring/groupthink in the interview results capture meeting (as well as other, more subtle, reasons to do it): Ask people to send you their recommendations and justifications before the meeting.
- Added 2022-10-19, edited 2022-10-23: In general, the items about ‘inaccurate or bad advice’ have common-sense arguments for and against them. (If anyone wants me to write any of them out, let me know.) It would come down to analysing Manager Tools’ (unfortunately non-public) data as well as the research you’re referring to. I would expect ‘traditional hiring’ when done right to work roughly equally well to whatever good EA orgs and leading tech firms do. My default expectation when people say that the ‘traditional way to do X is bad’ is that it actually works well when done right, but that it has a bad image because it got corrupted over time/is usually done mediocrely.
What links here?
- Be careful with (outsourcing) hiring by Richard Möhn (17 Oct 2022 20:30 UTC; 40 points)

Richard Möhn 14 Sep 2023 2:44 UTC
8 points
5 ∶ 0
on: Hiring: hacks + pitfalls for candidate evaluation
The outline structure makes this easy to skim. Thank you!

Richard Möhn 30 Dec 2022 5:39 UTC
8 points
7 ∶ 4
on: Givewell, and its hiring process, needs serious reform
Meta comment: There are many downvotes, but barely any comments. Feels a little uncivil. Certainly, the post has disagreeable points. But it’s useful input, too.

Richard Möhn 28 Dec 2022 22:08 UTC
8 points
1 ∶ 0
on: Things I didn’t feel that guilty about before getting involved in effective altruism
My experience: I’ve been living in the south of Japan for six years. It’s far from the most effective place to be. I gave up on being an independent researcher. I wanted financial stability and got a permanent work contract. If things go well, I’m going to have children soon. (Will MacAskill encourages people to have children in What We Owe the Future, by the way.) I don’t feel that guilty because I’m still a lot more effectively altruistic than the vast majority of people who would have the means.

Richard Möhn 18 Oct 2022 8:08 UTC
8 points
0 ∶ 0
in reply to: MichaelA’s comment on: Be careful with (outsourcing) hiring
They are the third and fourth row in the table, but the rows aren’t ordered by the validity column. When you order by the validity column, integrity tests are 8th and conscientiousness tests are 12th unless I’ve miscounted.

I admit that I cherry-picked this article, basically only looked at the validity numbers in the table, and don’t know anything else from that literature. This post has a wider view for those interested: https://forum.effectivealtruism.org/posts/j9JnmEZtgKcpPj8kz/hiring-how-to-do-it-better On the other hand, the validity table was excerpted on an 80,000 Hours page for years, which gives me some confidence by proxy. Also, the points of my post rely only lightly on this article.

Added 2022-10-19: It would be nice if you removed the bolding on the text that you found to be wrong. Otherwise the impatient reader is apt to miss the bracketed text above.
What links here?
- Be careful with (outsourcing) hiring by Richard Möhn (17 Oct 2022 20:30 UTC; 40 points)

Richard Möhn 19 Nov 2022 22:37 UTC
7 points
4 ∶ 0
on: Moderator Appreciation Thread
Thank you, moderators! I didn’t notice any trolls or spam, which I guess is evidence of your good work. :-)

Richard Möhn 1 Apr 2023 23:01 UTC
6 points
1 ∶ 0
on: Reflections on applications, rejections and feedback
Thanks for writing this analysis! I agree with a most of it. One other argument I’ve heard for not providing feedback after job applications is that it carries legal risk. What the risk is specifically, I don’t know – perhaps a candidate could sue you for discrimination if you inadvertently choose the wrong words? A way to mitigate this risk is the phrase ‘You didn’t demonstrate… [mastery of Python, facility in calming down angry customers, whatever]’. It avoids making statements about the candidate in general and instead focuses on their behaviour during the application process, which you’ve observed. [1] It can be addition to your points about how to give feedback.

[1] https://www.manager-tools.com/2013/07/you-did-not-demonstrate-part-1-hall-fame-guidance

Richard Möhn 18 Oct 2022 23:15 UTC
6 points
0 ∶ 0
in reply to: Kirsten’s comment on: Be careful with (outsourcing) hiring
It turns out that the article was mass-downvoted by throwaway accounts. So there weren’t many downvotes to find a reason for after all. Thanks for your comment anyway. It allowed me to partially address some weaknesses of my article.

Richard Möhn 4 Sep 2023 12:22 UTC
5 points
3 ∶ 0
on: Richard Möhn’s Quick takes
It’s long, but it looks like Rethink Priorities have put a lot of thought into this job posting: https://careers.rethinkpriorities.org/en/postings/8588ecc5-3e26-4086-bdb2-fe9a2eb43252

Richard Möhn 17 Nov 2022 8:20 UTC
5 points
3 ∶ 0
in reply to: WilliamKiely’s comment on: Kelsey Piper’s recent interview of SBF
Thanks for your concern about my karma! ;-) The difference between my comment and the reply is indeed surprising. I wouldn’t have noticed it if you hadn’t pointed it out. Here are other, more or less plausible, explanations:
- Most likely: My comment was a quick, abstract throwaway. I even avoided the word ‘suicide’ because I didn’t feel good about bringing it up explicitly. NeoMohist’s reply is more concrete, emotional and caring. I guess people resonate with that more and want to say with their upvote: ‘Yes, me worried too. Please keep an eye on him.’
- It was a quick, throwaway comment bringing up suicide. It carries bad vibes and doesn’t add much. (I guess on Facebook it’s harder/harsher to ‘downvote’ and people just decide not to react in that case, leaving you with the positive reactions.)
- NeoMohist has a cool username.
- EA Forum behaviour is deteriorating. (I was sometimes puzzled how things developed with my article on hiring and the comments on it. Of course, I’m biased for my article and comments.)
- Without the clarification (although I added it before all the karma accrued to the reply), one could misunderstand my comment as: ‘I’m only slightly worried because losing Sam wouldn’t be a big loss, given what was happened.’ (I feel bad for even spelling out this possible misinterpretation.)
- People read continuously until the end of the comment chain and vote only there, instead of going back and looking what else might deserve an upvote.

Richard Möhn

Be care­ful with (out­sourc­ing) hiring

Be careful with (outsourcing) hiring