A Two-Tier Model of Optimal Career Design
How I am scientifically designing my best strategy for living
I think all the time about my 'model of fulfilment', and the actions I need to do to live a maximally happy life. My first assertion is the obvious. Fulfilment is some function of a) personal comfort, and b) helping others. But finding out the best way to do this is hard. There are a lot of options, and here are common ‘default lines of questioning’ that I hear a lot from people
Should I follow my passion?
Should I be a charity worker when I'm young and then focus on family later on in life?
Should I always aim to take risks at doing good?
Should I try to make a lot of money first then donate all my money when I die?
Should I go on a gap year to discover myself?
A Two-Tiered Model as My Best Prior
My current best prior is a two-tiered model. The first tier consists of fulfilling personal needs, and only once that is under control, shift my focus into moving on to the second tier. The second tier consists of trying to gain fulfilment through outsized usefulness to society, via the taking of repeated risks. This is similar in spirit to Maslow's hierarchy of needs, but I don't think that it's necessary to have that many layers. I will explain why I think this structure works throughout the post.
Note, this post is also quite similar to the strategy by 80000 hours on ambition, which I strongly suggest you read as well.
But first, here is what this model is not. This model is different from just exclusively prioritising personal needs, like getting rich to enjoy one's wealth. It's also different to just focusing on being useful to society exclusively, say a charity worker or serial volunteer who is struggling to make ends meet.
More specifically, I also think that the order is important - for the second tier to work, the first tier must be in place.
So why two-tier?
My core assumption is that a significant chunk of outsized societal contribution comes from the small set of exceptional individuals that are unusually suited to their jobs, and who have taken risks to do something new.
Whilst I am still not completely certain of this claim, I have observed evidence of this through studying the history of the sciences. Fields have been revolutionised repeatedly by the same people with odd hunches think Einstein, von Neumann, Curie, Feynman, Witten, and Tao. And the reason why risk is involved is because outsized contributions need to stem from novelty, and so there is uncertainty around success. I highly suggest you read Simon Singh’s book on how Andrew Wiles approached Fermat’s last theorem.
A caveat to this point is that apparently we don't seem to have many prodigious people of this caliber anymore - which could just mean things have changed, and that discovery now necessarily requires lots of people. Science could just be a lot more efficient now, and Holden Karnofsky writes a great piece explaining the reasons why we haven’t had a Beethoven of modern times in a while.
Also, the opposite way works as well, when it comes to harm, only a small set of individuals are required to do outsized damage, like dictators - Hitler and Stalin.
And so, given that outsized contribution comes from people who are risk takers that are unusually suited to their domain, there are three things to optimise for
repeatedly risk taking
repeatedly trying to do build things
try to find something that you are unusually suited at
Notice how I say 'repeatedly'. I will admit to being selfish here. Ideally, I want be the one to stumble across a contribution myself. And so, like continually buying really cheap lottery tickets, the aim is to find ways to always just try again. And so, given that an action with high upside is risky, the best thing that you can do is to keep repeating cheap, risky actions with high upside. Traders would call this 'buying out the money calls'. By doing this, I am also implying that time, money and effort are not necessary constraints to making an outsized contribution to something.
Again, I'm relying on the assumption here that outsized contributions can come from exceptionally fit individuals with relatively little time and resources. Learning physics has taught me that time and effort are not necessarily the constraints to do amazing work. If we pour infinite amounts of resources in a poor strategy, we will get nowhere. The recent failures of the billion dollar upgrades of CERN to detect new physics at high energy levels is evidence of this. In fact, physicist Sabine Hossenfelder has repeatedly criticised this as a poor strategy - I suggest you read her book. And on the flip side, quantum mechanics, special relativity and general relativity where ideas sparked by basically one person - Einstein, in the span of five years.
So then, the question now is - how does one go about achieving the three things I mentioned risk taking, repeated trying, and unusual fit?
Well, this is the reason why the two-tiered model works. The first tier, having security and a safety net, allows for repeated risk taking, since your first basic needs are met.
To afford to be able to take the pain of failure, and fully take the plunge, a safety-net is required so that you can try again. The safety net allows one to make use of the law of large numbers - repeated risk-taking, which leads to eventually coming across an outsized contribution throughout many trials. Similar to the Silicon-Valley type strategy of failing fast, quickly and early.
For me the safety-net looks like this, which I admit is probably quite a high bar. In fact, I used to think I was risk averse, but now I just think that my situation and personal values calls
financial security, which, since I'm not from the UK and really can't afford to go back right now, is probably a higher buffer than most
having a house, which in London is quite expensive
starting a family, which again is quite expensive
being able to have free time
having the freedom to travel, which is expensive, but I've recently started to go bike touring which reduces the variable cost over the long term (despite having initial high fixed cost)
sending my children to a top school, which is just darn expensive!
I will say, that as an immigrant into the UK, that the bar for me to achieve this safety net has always felt higher that my local peers, at least mentally. For example, since I require a work visa, and visas require pretty high minimum salaries, there is pressure to maintain a high paying job. I also don't have a family home to resort to in the UK, which means I have to pay rent in London, which is a lot. Also, I do want to send my (future) kids to a really good educational institution, which again in the UK is just really expensive. Also I have to pay for flights back home to South-East Asia, which is, guess what, expensive!!!!
My strategy to achieve the safety-net in as efficient way as possible, without doing painful things, was to get into quantitative trading. But my main point is that there are quite a few other ways to do this, like working in tech or consulting, or working in government service where jobs are quite secure and tend to come with a decent pension package. As for quantitative trading, I enjoy it, the hours aren't bad, and it has some monetary upside if I perform well.
Once that is out of the way, then the next aim of the game is to find something that I could have an outsized contribution to, take some risk and then fail quickly to see if it works. I have found that the only way to do this is by reading lots of stuff and trying lots of stuff.
This is where I am now, and here are some of the things that I've tried over the past few years that haven't exactly worked out yet, but could've been great if something clicked !
Working with a computational chemistry lab in the UK to build spectroscopy predictions
Writing a book
Doing an art exhibition
Working with a lab in Malaysia to find ways to build cheaper graphene nanotech
Applying (and getting) a startup accelerator offer to try and make AI assistants
Working for a government agency in monetary policy
Trying to be a Youtuber on math
Trying to be a cycling influencer
Trying to be a substacker
Building and soldering chips in my house
Part of my strategy in doing this is doing ‘low cost trials’, which is where I try to do small scale projects in my free time to see whether something clicks and I can go further. This is what the Japanese author Haruki Murakami did when he wrote his first book - he worked on his bar, and then wrote his first novel in his free time to see if he could do it.
Doing these low cost trials for me often consists of writing blog posts, emailing people, coding and building what I think might be useful, and also doing minimal trust investigations on the topic, which is something I wrote about previously. The main thing I am optimising for when doing these low costs commitments is to see if
there are obvious ways that this could spiral into something that is potentially big
if I can resolve some key questions about what I am doing is actually impactful or not, and if I can resolve some key uncertainties
I am an unusually good fit for this type of work. In this case, I actually do find comparing myself to people a useful exercise, but adjusted for the amount of time an effort that I have put in. I want to feel like I am gliding whereas most others are wading, due to some edge that I have. For some examples where I have engineered some edge, check out this post.
Another reason why doing a two-tiered system works, is that you don’t necessarily have to do work in the bounds of constraints that other people have, which can be a catalyst for novelty. I wrote about this in a previous post here. I think this is best summarised by a quote by Richard Feynman:
So I have just one wish for you – the good luck to be somewhere where you are free to maintain the kind of integrity I have described, and where you do not feel forced by a need to maintain your position in the organization, or financial support, or so on, to lose your integrity. May you have that freedom.
There are also other ways which I’m trying to find upside potential, like talking to interesting people, cycling around the world in an attempt to make serendipidous connections, and just in general trying to keep myself open to doing things. Often times, I’ve decided to spend hours of my day just cold emailing people I am curious to talk to. It rarely works, but when it does, I’ve discovered a lot. In fact, it’s how I stumbled upon my current supervisor.
The end goal for me would be to do then transition into doing work that is uniquely suited to me, that I can have an outsized impact, but also pays me enough to maintain my safety-net requirements. The search continues!
Biorisk
Throughout trying these risky things, I am trying to keep in the back of my mind how they might be maximally impactful in reality. Currently, one thing I am looking into is trying to learn about neglected cause areas. And something that I’ve been using as a resource has been the 80000 hours’ content on neglected funding areas of existential risk problems.
Even though I am skeptical about a bunch of the claims written in this blog post, I like the way that the organisation thinks about it. In particular, I am quite interested in biological risks, how neglected it is as a cause area, and how I might contribute to it in an outsized way. But for now, I am quite uncertain.
Here are my uncertainties in order
What are the costs of the next devastating pandemic?
Is biosecurity actually a neglected funding area? Read this for what I think is the best breakdown so far.
Will I have outsized performance in any area of Catastrophic Risk work?
Will it energize me enough to work on this every day and transition to full time?
I am unsure of what the skills are required to do this work / level of technical proficiency / day to day.
Unsure of whether I'll enjoy the pace of policy research
In terms of topic questions, I’m also trying to think of models of potential threat from
malicious actors bioengineering synthetically harmful viruses
the fact that its easier to learn how to make harmful chemistry with LLMs
So currently, I am spending some time to do some minimal trust investigations on these topics. One thing I am focused on right now is actually figuring out if biorisk research is actually an underfunded area. I mean, the cynic in me thinks its plausable that countries invest in making bioweapons, and to do so, that they must invest in the appropriate safety measures to not harm their researchers. But overall, I’m not sure. I’m also applying to quick projects where I can maybe write an essay or post based on some hard data.
One thing that I’ve been looking at is DoD declassified budgets on recent spending. You can find the 2025 plan here, which goes into excrutiating detail on each of the project proposals the department wants to fund.
'The FY 2025 budget request of $1,656.7M enables the CBDP to translate strategic guidance into concrete capabilities tailored to the future threat. This budget request continues the ENBD efforts to modernize the Department’s biodefense capabilities to stay ahead of the threat.'
Seems like a small number!
Beautiful pixel art on the cover! Who is the artist?