Thank you for your response!   My concern is that cheap tests that only last a few weeks or months, will not provide accurate information about how good one is at a role.   I've been a software engineer almost two years now.   If I were to have only worked as a software engineer for six months and stopped and reflected on whether I had the potential to get very good at the role, I might have concluded that I had little potential based on my performance.   But now on month 19, I think my prospects are quite good.   There was  a really long onboarding and skill-building period that had to be done before I could really start contributing and determining how good I could get.   While there might be shorter onboarding periods in other jobs that are more oriented around soft skills - e.g. sales, consulting, marketing, etc., I imagine it will still take a long time to be sufficiently onboarded to be able to assess one's potential. 

However,  maybe there is a happy middle ground between the extremes of working on a role for 2 years and doing a small project for a few months.   I think the charity option is interesting - one could work for a charity in a role for a year or more to see if they'd be good at the skills involved in the role.  Probably a lot of roles - e.g. sales, marketing, accounting can be tested out this way.   But some roles like product management seem to be particularly hard to find in the volunteering space.  I've scoured Google for "volunteer product management" positions and only found 3 that seemed to be open to applications.  I interviewed for one of the three and it turned out be more of a project management role, where the volunteer had little agency/ownership.   

That said, maybe it's the case that the vast majority of skills can be  tested in volunteer roles. While it may be hard to find a volunteer product management position, it probably would be pretty feasible for a capable person to find roles that involved customer research, project management, web/mobile analytics, marketing, etc. at volunteer orgs, so one could potentially test out all the component skills used in product management separately.   In practice though, testing all these skills might be very difficult. 

Any thoughts for trying out jobs post-university?

I'm a 23 year old software engineer at a large tech company working out of NYC.  I've thought of the following ways to try out jobs:

1. Rotational programs  - Some tech companies offer programs where you can rotate roles every 6 months, like this one. But I've pretty much applied to all of them.
2. Full-time jobs - I could work full-time as a PM, UX Designer, university researcher etc. but this would probably require a 1-2yr commitment per job
3. Startup cofounder - As a cofounder I'd gain exposure to a lot of skills - sales, product, engineering etc. but I feel that I'd get such a small exposure to each function that it wouldn't really be useful

Any other ideas or thoughts?
I'm trying to figure out what sort of work I'm best at (which I imagine is where I'll have the most impact).

Out of curiosity, how long did you do your experiments in UX / Data Science / etc. for?   Maybe it would pay to try spending more time in these functions?   You could transition to data science and work for a year and half, then work for a year in UX for example (maybe do a bootcamp as a refresher / to build career capital).  Of course this is easier said than done - but I feel like it might take a long time before you can really assess how much a role aligns with your strengths, as it might take many months just to onboard to the role.   

I'm in my 20s working as a software engineer for a large US tech company and I hope to transition to some other roles for a few years if possible before committing to one role.  

Also, maybe it would be worth transitioning to a different company or team where you can feel that you work is having a greater impact on customers / on the world?   I feel like it can be a bit hard to feel your impact as a software engineer because usually you are usually not client facing.   But I imagine that your work might feel more impactful at a mission driven startup or if you are working as sales engineer and get to work with clients face to face periodically.