When I was starting out career-wise, I read a lot of advice from the EA community to the effect of “you’re bad at predicting what you’ll like, so it’s sensible to just try a few things and start gathering data.”
On balance I think this is smart, but it suffers from a very practical case of the problem of induction. What if, on a given career trajectory, your job will become very different after a certain amount of time in a way that is difficult to forsee when starting out?
A concrete example:
Aaron has been working as a programmer for a few years. For the first couple of years this was an extremely good fit: he had to assimilate a large amount of unfamiliar data, which is something he finds enjoyable. His own inexperience and the need to produce results also meant he needed to look for patterns and form rough, often inaccurate heuristics for how various programs worked to get anything done; a more systematic approach would have been too slow for his employer.
Now that he’s been in the industry for a few years, this no longer applies. Aaron has a good grasp of the fundamentals, and to make progress he needs to take a much more thorough and systematic approach to understanding the programs and systems he works with. This is a big enough change to the cognitive processes required of him that he’s gone from excelling in his job to being burnt out and likely needing a career switch.
Reflection:
The above example is of a good fit → bad fit switch over time due to the changing requirements of a job as you level up; it is also possible to make the same error in the other direction. Experimentation could cause you to de-select a path you initially find very difficult but that would become an extremely good fit after a few years.
To be clear, I still think experimentation is one of the best options for making career choices under high uncertainty, but I’m interested in how we can mitigate failure modes like the above. And I think a good first step here would be a knowledge-dump from people who have a) experienced changes to their role while remaining in a particular line of work, and b) think certain newcomers to the field should be aware that these changes will happen.
Request:
I would be excited to read comments from forum members including your job, the nature of a change you experienced in your job, the time period around which it happened, the kind(s) of individual you think would be significantly affected, and how/why you think they’d be affected for better or for worse.
Aaron’s might look like this:
Career: Programming Nature of change: Optimal cognitive processes for problem-solving given my knowledge/experience of the domain. Time period: After ~2 years of full-time programming. Who should consider this: People with strong feelings towards the amount of data consumption and nature of the problem-solving programming requires. Why it matters: If you enjoy consuming large amounts of information, or you enjoy solving problems with rough heuristics due to limited time and lack of familiarity with a complex problem space, be aware that both of these experiences may drop off significantly once you start to gain core competencies in programming. Equally, if there are aspects of programming you really like, but you find one or both of these features difficult, be aware that they may get a lot more manageable with time.
Conclusion:
I’m excited and hopeful about dodging more of these curveballs moving forwards; if you have experiences you can share, please share them!
Curveball Alert Request: How has your job changed over time?
Introduction:
When I was starting out career-wise, I read a lot of advice from the EA community to the effect of “you’re bad at predicting what you’ll like, so it’s sensible to just try a few things and start gathering data.”
On balance I think this is smart, but it suffers from a very practical case of the problem of induction. What if, on a given career trajectory, your job will become very different after a certain amount of time in a way that is difficult to forsee when starting out?
A concrete example:
Aaron has been working as a programmer for a few years. For the first couple of years this was an extremely good fit: he had to assimilate a large amount of unfamiliar data, which is something he finds enjoyable. His own inexperience and the need to produce results also meant he needed to look for patterns and form rough, often inaccurate heuristics for how various programs worked to get anything done; a more systematic approach would have been too slow for his employer.
Now that he’s been in the industry for a few years, this no longer applies. Aaron has a good grasp of the fundamentals, and to make progress he needs to take a much more thorough and systematic approach to understanding the programs and systems he works with. This is a big enough change to the cognitive processes required of him that he’s gone from excelling in his job to being burnt out and likely needing a career switch.
Reflection:
The above example is of a good fit → bad fit switch over time due to the changing requirements of a job as you level up; it is also possible to make the same error in the other direction. Experimentation could cause you to de-select a path you initially find very difficult but that would become an extremely good fit after a few years.
To be clear, I still think experimentation is one of the best options for making career choices under high uncertainty, but I’m interested in how we can mitigate failure modes like the above. And I think a good first step here would be a knowledge-dump from people who have a) experienced changes to their role while remaining in a particular line of work, and b) think certain newcomers to the field should be aware that these changes will happen.
Request:
I would be excited to read comments from forum members including your job, the nature of a change you experienced in your job, the time period around which it happened, the kind(s) of individual you think would be significantly affected, and how/why you think they’d be affected for better or for worse.
Aaron’s might look like this:
Career: Programming
Nature of change: Optimal cognitive processes for problem-solving given my knowledge/experience of the domain.
Time period: After ~2 years of full-time programming.
Who should consider this: People with strong feelings towards the amount of data consumption and nature of the problem-solving programming requires.
Why it matters: If you enjoy consuming large amounts of information, or you enjoy solving problems with rough heuristics due to limited time and lack of familiarity with a complex problem space, be aware that both of these experiences may drop off significantly once you start to gain core competencies in programming. Equally, if there are aspects of programming you really like, but you find one or both of these features difficult, be aware that they may get a lot more manageable with time.
Conclusion:
I’m excited and hopeful about dodging more of these curveballs moving forwards; if you have experiences you can share, please share them!