Interesting! I'm way out in the middle of nowhere, and experience suggests that the greatest benefits of intellectual co-location happen with physical co-location as well. I wonder if there would be interest in a program with some overlap across airbnb or farm stays, where one visits a spot out in the woods with decent internet but few distractions, and stays for a while (a week or two sounds like a plausible guess to start iterating from) with a host who assumes a metacognitive role in the project that one is working on. It seems quite appealing from a hosting perspective -- doing a short-term cognitive job-shadow role like that for an expert thinker would be deeply enriching, and hosting many thinkers over the years would build a fascinating expertise in pattern-matching between them, crafting an ontology of how folks in a given field get stuck and un-stuck, etc.
And I don't think I'm the only prospective host who prefers a remote location because dealing with strangers frequently (as one must to live richly in a city) gets exhausting, yet enjoys deeper small-group interaction when it's available. There's also a social dynamic where visiting someone in the middle of nowhere gives the host greater control over how time is used, since excursions outside the homestead cost more travel time and thus warrant more careful planning. This dynamic seems like it could be quite helpful if the host's primary priority is to advance the success of the guest's project.
I'd like to do either side of this! Which I say in public to have an opportunity to advertise that https://www.lesswrong.com/posts/MHqwi8kzwaWD8wEQc/would-you-like-me-to-debug-your-math remains open.
Probably the most powerful intervention I know of is to trade facilitation of emotional digestion and integration practices with a peer. The modality probably only matters a little, and so should be chosen for what's easiest to learn to facilitate. Focusing is a good start, I also like Core Transformation for going deeper once Focusing skills are good. It's a huge return on ~3 hours per week (90 minutes facilitating and being facilitated, in two sessions) IME.
Both people ideally learn from existing practitioners for a session or two, ideally they also review the written material or in the case of Focusing also try the audiobook. Then they simply try facilitating each other. The facilitator takes brief notes to help keep track of where they are in the other person's stack, but otherwise acts much as eg Gendlin acts in the audiobook.
I've been offering various flavours of this to selected people for the past few years (using the normally ill-advised free option, but I don't super need to earn money currently and it feels good to ask people to pay it forward and do good for the world), with pretty good reviews. I'm not super looking to expand this currently, but might be open to testing out more people in a month or three, depending on where priorities fall and whether I think the person is doing unusually good doom reducing work.
1I did a session yesterday with @moonlight, which went pretty well. I ended up consolidating some notes that seemed good to share with new assistants, and then he wrote the introduction he'd personally have preferred.
I generally work out of google docs that serve as shared-external-working-memory, with multiple tabs.
Moonlight's Intro
[Written by the first thinking assistant working with Ray, writing here what I’d have liked to read first]
Important things:
- Ray is currently sick, so put some effort into speaking more softly and slowly.
- There is no interview or anything similar, you’ll begin assisting him straight away.
- By default, just watch him work (coding/planning/writing/operations), and occasionally give signs you’re still attentive, without interrupting.
- Write moment to moment observations which feel useful to you, as well as general thoughts, down in the Assistant Notes tab. This helps you feel more proactively involved and makes you focused on noticing patterns and ways in which you could be more useful as an assistant.
- The Journal tab is for his plans and thoughts about what to generally do. Read it as an overview.
- This Context tab is for generally useful information about what you should do and about relevant strategies and knowledge Ray has in mind. Reading this helps you get a more comprehensive view on what his ideal workflow looks like, and what your ideal contributions look like.
For the structure of this document:
- Collapse sections when reading. It helps traverse the document.
- This part has my thoughts for onboarding. Under it, you can find Ray’s onboarding section. Read these two first.
- The “Ray Facts” section has important information about logistics and operations. Currently it has only his work location. [edited out in this comment]
- In “Ray’s Metacognitive Engine” and below, you can find the strategies and knowledge I’ve mentioned above. You can read these after, they’re not mandatory at the very start.
Ray’s First Draft Intro Materials
Strategic Overview
Goal: End the acute risk period, and ensure a flourishing human future.
I’ve recently finished a bunch of grieving necessary to say “all right I’m ready to just level up into an Elon-Musk-but-with-empathy-and-cyborg-tools type”, as well as the minimum necessary pieces of a cognitive engine that (I think) is capable of doing so).
I want to be growing in capacity at an exponential rate, both in terms of my personal resources, and the resources available to the x-risk ecosystem that are accomplishing things I think need accomplishing.
This means having a number of resources that are compounding, that are synergistic, which include:
- Money (either mine, or ability to spend Lightcone’s)
- Skills
- Meta personal skills, like ability to learn, and understand things, or be strategic
- Meta interpersonal skills, such as the ability to outsource labor or make use of assistants,
- Object level skills like programming, UI design, Event running
- Ability to work with employees who can take on tasks I want done
- Capital
- Relationships with people I work well with
- Tools I can re-use
Things I actually do most days:
- Coding on LessWrong
- Coding on random other projects
- Planning my Cybercognition Agenda, which includes workshops, cybernetic tools, and upskilling people around me.
- UI design, trying to figure out important complex things I want people to interact with in a way that feels simple to them.
- Thinking strategically about what needs to be done next
Instructions for Thinking Assistants
Things I would like you to do:
- By default, be quiet and attentive and just help me focus by being a real human who’s staring at me
- Develop skills for tackling sort of arbitrary ops or research or coding tasks, such that I can outsource small things to you.
- Advice
- This is tricky because I have a good enough model of myself that a lot of advice isn’t that helpful. It’s still useful to have my blindspots pointed out. But, if I interrupt you (either with words or with a hand gesture) that probably means I want to move on to a different thread. (Ideally, you feel comfortable bringing up ideas, with no hard feelings if it doesn’t work out)
I would like to end up with a series of if-then habits you can help me execute. I will mostly write these myself, but as you get to know me well enough to say useful things, you can make suggested-edits
From “Hire or become a Thinking Assistant”
- By default, be quietly but visibly attentive.
- Every now and then (~5-10 minutes, or when I look actively distracted), briefly check in (where if I'm in-the-zone, this might just be a brief "Are you focused on what you mean to be?" from them, and a nod or "yeah" from me).
- When I need to think something through, they rubber duck (i.e. listen as I talk out loud about it, and ask clarifying questions)
- Build a model of my thought process (partly by me explaining it to them, partly by observing, partly by asking questions)
- Ideally, notice when my thought process seems confused/disoriented/inefficient.
- Ideally, have a large repertoire of cognitive tools they can suggest if I seem to be missing them.
- Intelligent enough that they can pretty easily understand the gist of what I'm working on.
- Ability to pick things up from context so I don't need to explain things in too much detail.
- Ideally, when my bottlenecks are emotional, also be at least fairly emotionally attuned (i.e. project a vibe that helps me worth through it, or at least doesn't add extra friction or emotional labor demands from me), and ideally, basically be a competent therapist.
- In general, own the metacognition. i.e. be taking responsibility for keeping track of things, both on a minute-to-minute timescale, and the day-to-day or week-to-week timescale.
- Ability to get out of the way / quickly drop things if it doesn't turn out to be what I need, without it being a big deal.
There are also important outside-the-container skillsets, such as:
- Be responsive in communication, so that it's easy to schedule with them. If it's too much of a pain to schedule, it kinda defeats the point.
- Potentially: proactively check in remotely during periods where I'm not actively hiring them. i.e. be a professional accountability buddy, maybe paid some base rate to briefly check in each day, with the ability to upsell into "okay today is a day that requires bigger metacognitive guns than Raemon has at the moment")
Even the minimum bar (i.e. "attentive body double") here is a surprisingly skilled position. It requires gentleness/unobtrusiveness, attentiveness, a good vibe.
The skill ceiling, meanwhile, seems quite high. The most skilled versions of this are the sort of therapist or executive coach who would charge hundreds of dollars an hour. The sort of person who is really good at this tends to quickly find their ambitions outgrowing the role (same with good executive assistants, unfortunately).
Pitfalls
Common problems I've run into:
- Having trouble scheduling with people. If you want to specialize in this role, it's often important for people to contact you on a short timeline (i.e. I might notice I'm in a brainfoggy state and want someone to assist me like right now, or tomorrow), so, having a communication channel you check regularly so people can ping you about a job.
- Asking questions in a way that is annoying instead of helpful. Since the point is to be giving me more time, if I have to spend too much time explaining the situation to someone, it undoes the value of it. This requires either them being good at picking things up quickly without much explanation, or good at reading nonverbal cues that the current thread isn't worth it and we should move on.
- Spending too much time on unhelpful advice. Sometimes an assistant will have ideas that don't work out, and maybe push them more than appropriate. There's a delicate balance here because sometimes I am being avoidant or something and need advice outside of my usual wheelhouse, but generally if advice isn't feeling helpful, I think the assistant should back off and observe more and try to have a few other hypotheses about what to suggest if they feel that the assistee is missing something.
- Navigating weird dynamics around "having someone entirely optimized to help another person." Having this run smoothly, in a net helpful way, means having to actually be prioritizing my needs/goals in a way that would normally be pretty rude. If I constantly feel like there's social awkwardness / wariness about whether I'm making them feel bad, the whole thing is probably net negative. I think doing a good job of navigating this requires some nuance/emotional-skill on both parties, in terms of striking a vibe where it feels like you are productively collaborating.
- (I think this likely works best when the person is really actively interested in the job "be a thinking assistant", as opposed to something they're doing because they haven't gotten traction on their real goals).
Ray’s Metacognitive Engine
- Twice a day, asking “what is the most important thing I could be working on and why aren’t I on track to deal with it?”
- you probably want a more specific question (“important thing” is too vague). Three example specific questions (but, don’t be a slave to any specific operationalization)
- what is the most important uncertainty I could be reducing, and how can I reduce it fastest?
- what’s the most important resource bottleneck I can gain, or contribute to the ecosystem, and would gain me that resource the fastest?
- what’s the most important goal I’m backchaining from?
- you probably want a more specific question (“important thing” is too vague). Three example specific questions (but, don’t be a slave to any specific operationalization)
- Have a mechanism to iterate on your habits that you use every day, and frequently update in response to new information
- for me, this is daily prompts and weekly prompts, which are:
- optimized for being the efficient metacognition I obviously want to do each day
- include one skill that I want to level up in, that I can do in the morning as part of the meta-orienting (such as operationalizing predictions, or “think it faster”, or whatever specific thing I want to learn to attend to or execute better right now)
- for me, this is daily prompts and weekly prompts, which are:
- The five requirements each fortnight:
- be backchaining
- from the most important goals
- be forward chaining
- through tractable things that compound
- ship something
- to users every fortnight
- be wholesome
- (that is, do not minmax in a way that will predictably fail later)
- spend 10% on meta (more if you’re Ray in particular but not during working hours. During working hours on workdays, meta should pay for itself within a week)
- be backchaining
- Correlates:
- have a clear, written model of what you’re backchaining from
- have a clear, written model of how you’re compounding
- The general problem solving approach:
- breadth first
- identify cruxes
- connect inner-sim to cruxes / predictions
- follow your heart
- see how your predictions went
- Random ass skills
- napping
- managing working memory, innovating and applying on working memory tools
- grieving
- Generalizing
Skill I’m working on that hasn’t paid off yet but I believe in:
- At least once a day or so, when you notice a mistake or surprise, spent a couple minutes asking “how could I have thought that faster” (and periodically do deeper dives)
- each day/week, figure out what you’re confused or predictably going to tackle in a dumb way, and think in advance about how to be smart about it the first time
I also have found I am way more productive when I can do something like this, and kind of want to figure out how to do more of it. Some things that make it harder than it seems like it should be:
- NDAs and such
- the things I most need help thinking about are usually also ones that it feels very vulnerable/kind of scary to bring someone else into
- and/or are ones where domain knowledge is important, such that I'd want to ideally work with someone who knows stuff about it
- general feeling of privacy/hesitation putting another human in my workflows because my workflows feel very personal (...in part because of things like ~shame around being less productive than I'd like, which is a kind of silly self-sustaining cycle but not necessarily trivial to exit)
some ways I work around this are -
- coworking with friends, with work pomos and break periods where we talk about how things are going; this is an equal relationship and not one where we can get very far into the weeds on each other's work usually, but it helps a lot to be in a shared work zone & to have explicit social ritual around talking through how things are going, which often leads to noticing possible improvements to strategy. extra good if we are all working on similar stuff, though not required
- text channels for narrating my thought process, privately or to an occasional audience (or Google docs for same but with more structure)
- if I keep being stuck on thinking about a given thing, talk to a friend about it
- identify specific friends who are well placed to help me with specific projects & invite them to work on that specific project together for an hour or day
- effort-trading where a friend and I help each other with projects on different days
It would also be nice to be able to pay for this as a service but I haven't quite been able to convince myself to try any of this with a stranger! very likely I'd benefit from more highly prioritizing attempting to experiment with versions of this, though.
I'd sort of naively guess doing it with a stranger (esp. one not even in your circles) would be easier on the "feeling private/anxious about your productivity" – does that feel like it wouldn't work?
no it feels scarier! I think if I'm interacting with a real live human being in person I basically always instinctively worry about what they think of me even if there's no strong reason to, and higher uncertainty about what they think of me is more worry-causing; with friends I can somewhat lean on "well they are continuing to be friends with me so they must not be judging me too badly", and also friends have often disclosed similar vulnerable things to me which makes it easier (I am somewhat more hesitant to share productivity details with friends who I feel are way more productive than me). also "entirely outside my circles" is likely to come with "high inferential gap about various stuff I care about". I don't think this all is definitely insurmountable but surmounting it is a currently slightly mysterious first step
(I think "figure out how to tolerate talking to an LLM" might be an easier inroad actually, though that's differently aversive for me)
1DM me if you're interested.
I, too am quite interested in trialing more people for roles on this spectrum.
FYI: I'm hiring for basically a thinking assistant, right now, for I expect 5 to 10 hours a week. Pay depending on skill-level. Open to in-person or remote.
If you're really good, I'll recommend you to other people who I want boosted, and I speculate that this could easily turn into a full time role.
If you're interested or maybe interested, DM me. I'll send you my current writeup of what I'm looking for (I would prefer not to post that publicly quite yet), and if you're still interested, we can do a work trial.
However, fair warning: I've tried various version of hiring people to support my metacognition over the past 5 years, and so far none of them have worked well enough that it was worth continuing. I've learned a bit about what I need each time, but a lot of this will probably come down to personal fit.
I've been testing this with @Épiphanie Gédéon for a few months now, and it's really, really good for doing more work that's intellectually challenging. In my opinion, the most important sentence in the post is the fact that it doesn't help that much during peak performance moments, but we’re not at our peak that often. And so, it's super important. It’s really a big productivity boost, especially when doing cognitively demanding tasks or things we struggle to "eat the frog". So, I highly recommend it.
But the person involved definitely needs to be pretty intelligent to keep up and to make recommendations that aren’t useless. Sometimes, it can feel more like co-working, there are quite a few different ways it can work, more or less passive/active. But overall, generally speaking, we recommend trying it for at least a few days.
It took me quite a while to take the plunge because there's a social aspect—this kind of thing isn’t very common in France. It’s not considered a real job. Although, honestly, it should be a real job for intellectual professions, in my opinion. And it’s not an easy job.
Every now and then (~5-10 minutes, or when I look actively distracted), briefly check in (where if I'm in-the-zone, this might just be a brief "Are you focused on what you mean to be?" from them, and a nod or "yeah" from me).
Some other prompts I use when being a [high-effort body double / low-effort metacognitive assistant / rubber duck]:
- What are you doing?
- What's your goal?
- Or: what's your goal for the next n minutes?
- Or: what should be your goal?
- Are you stuck?
- Follow-ups if they're stuck:
- what should you do?
- can I help?
- have you considered asking someone for help?
- If I don't know who could help, this is more like prompting them to figure out who could help; if I know the manager/colleague/friend who they should ask, I might use that person's name
- Follow-ups if they're stuck:
- Maybe you should x
- If someone else was in your position, what would you advise them to do?
I like all these questions. "Maybe you should X" is least likely to be helpful but still fine so long as "nah" wraps up the thread quickly and we move on. The first three are usually helpful (at least filtered for assistants who are asking them fairly thoughtfully)
Another thought, though I don't actually have any experience with this, but mostly doing attentive silent listening/observing might also be useful for learning how the other person is doing research.
Like, if it seems boring to just observe and occasionally say sth, try to better predict how the person will think or so.
I'm interested in variants of this from both sides. Feel free to shoot me a DM and let's see if we can set something up.
I haven't had a good label to put on things like this but I've gravitated towards similar ways of work over the last 10-20 years, and I've very often found very good performance boosting effects, especially where compatibility and trust could be achieved.
I haven't actually tried this, but recently heard about focusbuddy.ai, which might be a useful ai assistant in this space.
Then, since I've done the upfront work of thinking through my own metacognitive practices, the assistant only has to track in the moment what situation I'm in, and basically follow a flowchart I might be too tunnel-visioned to handle myself.
In the past I have literally used flowcharts for this, including very simple "choose your own adventure" templates in roam.
The root node is just "something feels off, or something", and then the template would guide me through a series of diagnostic questions, leading me to root nodes with checklists of very specific next actions depending on my state.
The fact that you have and are using flowcharts for that use is very validating to me, because I've been trying to create my own special flowcharts to guide me through diagnostic questions on a wide range of situations for about 6 months down.
Are you willing or able to share any of yours? Or at the very least what observations you've made about the ones you use the most or are most effective? (Obviously different courses for different horses/adjust the seat - everyone will have different flowcharts depending on their own meta-cognitive bottlenecks)
Mine has gone through many iterations. The most most expansive one is it lists different interrogatives "Should I..." "Why do I..." "How can/should I..." and suggests what I should be asking instead. For example "Why do I always..." should be replaced with "Oh yeah, name three times this happened?" which itself begs the problem statement questions - Why did you expect that to work (How how confident were you/how surprised when it didn't)? How did it differ from your expectations? How did you react (and why did you react in that way)?
The most useful one is a cheatsheet of how to edit videos, with stuff like "Cut at least one frame after the dialogue/vocals comes in", "if an edit feels sloppy, consider grouping B-roll by location rather than theme/motif". It's not really a flowchart in that there's rarely branching paths like the question one.
Does this, at least structurally or implementation wise resemble your most effective flow-charts?
Okay a few people have DMd me, and I'm feeling some kind of vague friction that feels currently on track to be a dealbreaker so let's think that through here.
Problems:
- I can't tell offhand who's good at this, and while I think this is something someone with little experience could turn out to be good at, they often won't be, and it's kind of costly to spend a slot on them, especially if I really need someone competent at it.
- I often need someone "right now", and need a way to contact a bunch of people quickly, such that most of them will get the message and one of them will reply quickly, in a way that isn't too annoying for them but works.
I have a vision of a whole-ass website dedicated to facilitating this but right now want a quick hacky solution.
A group DM would work, but that feels like it'll produce weird competitive dynamics with who replies first but maybe isn't as good as the person who replies second.
DMing a bunch of people individually I guess is fine but but then I need to go find them.
A requirement for everyone participating as an assistant is that they have a way of being contacted that they'll respond to quickly.
I think more people (x-risk researchers in particular) should consider becoming (and hiring) metacognitive assistants
Why do you think x-risk researchers make particularly good metacognitive assistants? I would guess the opposite - that they are more interested in IC / non-assistant-like work?
Hazarding a guess from the frame of 'having the most impact' and not of 'doing the most interesting thing':
- It might help a lot if a metacognitive assistant already has a lot of context on the work
- If you think someone else is doing better work than you and you can 2x them, that's better than doing your individual work. (And if instead you can 3x or 4x people...)
I actually meant to say "x-risk focused individuals" there (not particularly researchers), and yes was coming from the impact side of things. (i.e. if you care about x-risk, one of the options available to you is to becoming a thinking assistant).
1I do things like this at times with my teams.
Important things:
-
Don't think you need to solve the actual problem for them
-
Do solve 'friction' for them as much as possible
-
Do feel free to look up other sources so you can offer more perspective and to take off the load of having to find relevant info
-
positive energy, attentive etc
-
if they're functioning well just watch and listen while being interested and unobtrusive, at most very minor inputs if you're pretty sure it'll be helpful
If stuck at a crossroads ask them how long they think each path will take/ how hard it'll be, and give them feedback if you think they're wrong. Help them start working on one, people can get stuck for longer than it would take to actually do one option.
- if lost, methodically go through the different areas where the issue could be and methodically go through all the directions they could take for each area and in general. You don't need to think these up, but keep track of them and help guide towards picking apart the problem and solution spaces. This takes some mental load off.
(Quick note to people DMing me, I'm doing holidays right now and will followup in a week or so. I won't necessarily have slots/need for everyone expressing interest)
I’m pretty optimistic that you could build limited-scope versions of this based on AI. Eg a Vscode plugin that observes you writing code, or a browser extension that observes you writing in Google Docs. Also plausible that the AI thinking assistant would be better than median human, or would soon reach that level.
I think the minimum viable "attentive body double + rubber duck" is something AI could implement right now. ChatGPT's voice mode would basically be fine at this if it:
In fact, there is an already existing implementation of this. James Campbell's ProctorAI allows an multimodal LLM to send "personalized messages" based on screenshots, and Campbell claims that:
In my experience, I tend not to break the rules because I can intuitively feel the AI watching me--just like how test-takers are much less likely to cheat when they can feel the proctor of an exam watching them.
It seems mostly successful at minimizing procrastination. I have not tried it though. The LLM also only observes the computer; you'll need to change the code if you want the LLM to observe (and hopefully minimize procrastination on) non-computer activities.
Staying on task, and being only accountable to myself are difficult for me, usually. The idea of a body double would, in my case, be useful. Having enjoyed teaching online for a few years I know often people want a tutor in person. Not for me any more. I'd be totally up for a virtual version of supporting another's work.
Unless you’re paying gratuitously, the only people who would reliably be interested in doing this would be underqualified randoms. Expect all benefit to be counteracted by the time it takes to get into a productive rhythm, at which point they’ll likely churn in a matter of weeks anyway.
Counterexample: financially self-sufficient individual who is curious about the work that the thinker is doing, and wants to learn more of how it is done.
1Oops! I only realized in your reply that you're considering "reliability" the load-bearing element. Yes, the hiring pipeline will look like a background noise of consistent interest from the unqualified, and sporadic hits from excellent candidates. You're approaching it from the perspective that the background noise of incompetents is the more important part, whereas I think that the availability of an adequate candidate eventually is the important part.
I think this because basically anywhere that hires can reliably find unqualified applicants. For a role where people stay in the job for 6 months, for instance, you only need to find a suitable replacement once every 6 months... so "reliably" being able to find an excellent candidate every day seems simply irrelevant.
I imagined "FocusMate + TaskRabbit" specifically to address this issue.
Three types of workers I'm imagining here:
- People who are reasonable skilled types, but who are youngish and haven't landed a job yet.
- People who actively like doing this sort of work and are good at it
- People who have trouble getting/keeping a fulltime job for various reasons (which would land them in the "unreliable" sector), but... it's FocusMate/TaskRabbit, they don't need to be reliable all the time, there just needs to be one of them online who responds to you within a few hours, who is at least reasonably competent when they're sitting down and paying attention.
And then there are reviews (which I somehow UI design to elicit honest reactions, rather than just slapping a 0-5 stars rating which everyone feels obligated to rate "5" all the time unless something was actively wrong"), and they have profiles about what they think they're good at and what others thought they were good at.
(where an expectation is, if you don't have active endorsementss, if you haven't yet been rated you will probably charge a low rate)
Meanwhile if you're actively good and actively reliable, people can "favorite" you and work out deals where you commit to some schedule.
Depth of specialization to the individual is an interesting question. I suspect that if this was a mature field, we'd have names for distinct subtypes of assistant skillset -- like how an android app dev isn't quite the same as an ios app dev, although often one person can do whichever skillset a situation demands.
I suspect that low-skill candidates would gravitate toward one assistance subtype or another, and lack of skill would show up in their inability to identify which subtype a situation calls for and then adapt to it. But on taskrabbit, we don't need the same tasker to be good at picking up groceries and also building furniture, as long as we're clear enough about which task we're asking for...
It’s true any job can find unqualified applicants. What I’m saying is that this in particular relies on an untenably small niche of feasible candidates that will take an enormous amount of time to find/filter through on average.
Sure, you might get lucky immediately, but without a reliable way to find the “independently wealthy guy who’s an intellectual and is sufficiently curious about you specifically that he wants to sit silently and watch you for 8 hours a day for a nominal fee”, your recruitment time will, on average, be very long, especially in comparison to what would likely be a very short average tenure given the many countervailing opportunities that would be presented to such a candidate.
Yes, it’s possible in principle to articulate the perfect candidate, but my point is more about real-world feasibility.
Of the posts I've delayed writing for years, I maybe regret this one the most.
I think more people (x-risk focused people in particular) should consider becoming (and hiring) metacognitive assistants. This is the single largest performance boost I know of – hire someone to sit with you and help you think. It doesn't help me (much) when I'm at my peak, but I'm not at my peak most of the time.
There are four types of assistants I'm tracking so far:
Body doubles just sit in the room with you, periodically looking at your screen, and maybe saying "hey, do you endorse being on facebook?". They're a kind of brute force willpower aid. The person I know who uses them the most (Alex Altair) has them just sit in the same room (I believe while doing pomodoros, each of them working on different things). He guesses that they 2x his productivity (which is around what I've gotten)
A metacognitive assistant is a step beyond, where they are dedicating their attention to you, noticing when you are getting stuck, and gently intervening. (I assume people vary in how they like to be intervened on, but for people doing nuanced cognitive work, I think not disrupting someone's thought process is very important. You need to feel safe with a metacognitive assistant). My experience is that this is a 1.5x to 2x multiplier on my output.
The next two types are both more involved than Metacognitive Assistants, but in different ways.
Tutors pay attention to you, but are particularly modeling how you are approaching a particular skill (programming, math, etc). They notice when you seem to be tackling the skill in a confused or inefficient way, and ask questions about your thought process so as to figure out what subskills or concepts you need to develop.
Partners or apprentices are full on "pairing" – they actively collaborate with you on your task. Hiring a partner/apprentice is a very hard task, it requires tons of chemistry and intellectual compatibility, so it's not really a shortcut to anything, but if you find the right person it seems great.
(John Wentworth says his research partner David Lorell multiplied his productivity by 3x, largely by raising John's floor performance. His earlier estimates were higher, and he says the current 3x takes into account that the trend of value-estimate has been downward. He does flag that the reduction-in-value-estimate included "dealing with some burnout" at times when he ended up pushing himself harder than he'd have naturally done if working on his own. He's since iterated on how to deal with that).
This post is mostly focused on Metacognitive Assistants, because I think they a) require some upfront investment to turn into a functioning niche of the rationalsphere (moreso than body doubles), b) feel achievable to scale up (whereas Tutors/Partners are both pretty advanced roles).
Pricing here varies wildly. I believe Alex Altair mostly hires UC Berkeley grad students for ~$15/hr, I've worked with people in more dedicated Metacognitive Assistant roles for $40-$80/hr depending on circumstances. Research assistants and tutors are probably much more bespoke.
Executive Assistants
I'm contrasting "Thinking Assistants" with "Executive Assistants." They do involve many of the same skillsets. I see executive assistants' job as a) handling your general metacognition across all the domains other than your core competency, and often handling various other personal-or-professional tasks that free up your time to focus on your core competency.
I think executive assistants are also great, and maybe they should blend with the Thinking Assistant role, since you realistically don't need a Thinking Assistant all the time and do need this other stuff dealt with and they probably collectively are worth one fulltime hire. But it is a different job.
Core Skills of a Metacognitive Assistant
I assume people will vary in what works for them. But, what I want out of a Thinking Assistant is:
There are also important outside-the-container skillsets, such as:
Even the minimum bar (i.e. "attentive body double") here is a surprisingly skilled position. It requires gentleness/unobtrusiveness, attentiveness, a good vibe.
A thing that feels a bit silly to me is that this isn't something I've been able to make work very well at Lightcone with other Lightcone employees. Sometimes we actively pair on tasks and that works well. But, our hiring process sort of filters for ornery opinionatedness, which is kinda the opposite of what you want here. I think even the simplest version of this is a specialized role.
The skill ceiling, meanwhile, seems quite high. The most skilled versions of this are the sort of therapist or executive coach who would charge hundreds of dollars an hour. The sort of person who is really good at this tends to quickly find their ambitions outgrowing the role (same with good executive assistants, unfortunately).
Pitfalls
Common problems I've run into:
Optimizing for (not-particularly skilled) Metacognitive Assistance
I've worked with people who were actively skilled at Thinking Assistance, and one person for whom it wasn't really their main thing, just a job.
One way I got more mileage out of the not-as-skilled person was to do upfront work of assembling a list of cognitive situations + habits. i.e:
etc.
Then, since I've done the upfront work of thinking through my own metacognitive practices, the assistant only has to track in the moment what situation I'm in, and basically follow a flowchart I might be too tunnel-visioned to handle myself.
Automated AI Assistants?
Like many professions, AI is probably going to automate this pretty soon. I think the minimum viable "attentive body double + rubber duck" is something AI could implement right now. ChatGPT's voice mode would basically be fine at this if it:
Presumably people are working on this somewhere. I might go ahead and build my own version of it since I expect to eventually want highly customized cyborg tooling for myself, and since AI is dropping the cost of developing apps from scratch. But, I expect the market to figure it out sooner or later.
This establishes a pretty solid floor in quality. But, since part of the active ingredient here is "a real human is paying attention to you and will hold you accountable with a bit of their human soul", I expect there to continue being at least some benefit to having a real human. (I think there will be some minimum bar of attentiveness + unobtrusiveness + able-to-follow that a human will need, to be worth using over an AI, once the AI is basically working)
Trialing People + Matchmaking
For the immediate future, I'd like to trial more people at cognitively assisting me, explicitly with a goal of being able to matchmake them with other people if appropriate. DM me if you're interested.
I also generally recommend other people trying experimenting with this in an ad-hoc way and writing up their experiences.
Focusmate + TaskRabbit?
It'd be nice to have a scalable talent pipeline for this, that matchmakes people with assistants.
Because of the combination of:
I think the natural vehicle here is a matchmaking site that's similar to FocusMate (which pairs people for coworking) but more like you're hiring skilled labor. I can imagine something where people list different skills and rates, and get ratings based on how helpful they've been.
Hypothetically this could be a very openended public-facing commercial website. I do personally feel like for a lot of work in the x-risk space it helps a lot to have someone in sync about my strategic frame and would feel more friction working with a more random general population person.
Aligning Incentives
An obvious idea that might occur to you is "Provide metacognitive assistance for free, to people you think are doing good work." I don't think this is a good idea longterm – I think it's a recipe for people ending up undervalued, as people model the cost as "free" rather than "subsidized." It also might turn into some kind of Lost Purposes Appendage where nobody knows how to evaluate either the research or the thinking-assistance and it gets propped up (or not) depending on how flush-with-funding the EAcosystem is this particular year.
I feel more optimistic about "the ecosystem overall figures out how much work various people's work is worth via various evaluation / grantmaking processes", and then people pay for metacognitive assistance if it's actually worth it.
Overall, this is one of the highest effect sizes I know of for productivity (up there with "get medication for your depression", "get a more motivating job" and "get enough sleep"). It is admittedly not cheap – $800/week at the cheap end if fulltime, and sort of unboundedly expensive at the higher end. (Modulo "maybe someone can build a good AI for this").
If you go this route – remember to keep track of whether you're overworking yourself. My current model is most people can in fact work more hours than they can motivate themselves to while working alone, but John's and my experience is that it's at least possible to overdo it if you're not careful.