Hide table of contents

This is the concluding post in a sequence called “Experiments in Local Communitybuilding”. This post specifically two previous ones called  “Consider splitting your fellowships into two parts” and “Onboarding students to EA/AIS in 4 days”, so to get the right context I would recommend reading those first.

TLDR: If you split your fellowships you can run more of them per semester by, for example, running a weekly one and then an intensive one during fall break, and then joining the two cohorts together for the second half.

Background assumption

Before starting, I thought I would make the following explicit: This post operates on the background assumption that high-fidelity reading groups and fellowships are a good thing, and generally speaking, we want more people doing more of them. Of course, there are lots of reasonable ways to criticize reading group-style community building (e.g. see here and here) and I have my own reservations as well. However, I think that reading groups can be great and they are here to stay for the foreseeable future in terms of EA/AIS field building, so we should explore how to make more of them or run them more effectively.

How and why

If you think running intro to EA/AIS fellowships is valuable, then here is how you can run 2 (or even 3) of them per semester or during the summer break. Combining the lessons from here and here, you can use the following structure for your fellowships below. Unfortunately, the picture is too big for the page, so you will have to click here for the guide. You can read about the structure there, and for the rest of the post, I will just outline some caveats.

Don't ask why I thought making a timeline would be a good idea on Mindjet manager. It wasn't. Click for the guide here.

Some general caveats

Running one fellowship is already plenty of operations work, and while running two doesn’t have to be twice as much, I would mainly recommend this for teams that have run a couple of weekly fellowships already, and/or have an experienced operations person.

An argument to consider against this approach is that fellowships should be something special, perhaps happening only twice per year. I think this has some merit, but in itself not a strong enough reason (for me) not to do them more often.

At first glance, it seems that the decision of whether to run 1 or more fellowships should depend on:

  • If you can get more people deeply engaged with your course’s content (which is already a good thing if you agree with my background assumption)
  • while having to do less work / engaged person finishing the fellowships
    • This should be possible if e.g.
      • you are combining a weekly and an intensive fellowship, see the 4th point of the guide here)
      • or if you have some effort invested into fellowship 1, but then can mostly copy over for fellowship 2 (e.g. you have to spend time figuring out how to market effectively, but then can mostly copy over your existing method for future rounds)

However, even if it’s true that you can onboard more promising people/units of effort by running more fellowships per semester, you should still think carefully about the counterfactual value of your time, as even if you get some efficiency gains, you will spend a larger fraction of your time on fellowships as opposed to other activities you might pursue as a communitybuilder.

Some personal caveats

I also just want to highlight that while we have run 2 fellowships per semester before, it was a combination of doing a pre-semester intensive fellowship and the usual weekly fellowship, as opposed to what I outline as the ideal(?) approach of combining the weekly and intensive fellowships for sessions 5-8. That’s also the version that is hardest to run, as you will have to make sure your weekly fellowship groups are on time for the intensive fellowships to join for the second half. I will make sure to update this post once we have run different iterations.

2024 May Update: We have run a version where we combined the weekly approach with an intensive fellowship ran during spring break, it went relatively well!


I think fellowships are great and we want more of them, as long as we are able to attract intellectually curious people. In case you want to try this approach feel free to shoot me a message, but also make sure to coordinate with your mentor!

Thanks to Milán Alexy for reviewing this post!

Sorted by Click to highlight new comments since:

Executive summary: You can run multiple fellowships per semester by splitting them into parts, like a weekly fellowship plus an intensive one, to onboard more people while minimizing organizational work.

Key points:

  1. Running multiple fellowships per semester can engage more people with less work if you reuse materials or combine fellowship types.
  2. A good approach is a weekly fellowship plus an intensive one joined together for the second half.
  3. This takes coordination and experience running fellowships, so gauge your team's capacity first.
  4. Consider whether more frequent fellowships reduce their specialness and your time for other tasks.
  5. The author hasn't run the ideal combination yet, but will update once iterations happen.



This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.

More from gergo
Curated and popular this week
Relevant opportunities