The question posted on Reddit is earnest, and it reveals a common anxiety: the belief that a fellowship invitation depends on a checklist of prestige markers, university name, publication count, famous recommenders. We think that framing misses the point. The ETH AI Fellowship does not exist to reward a résumé. It exists to identify people who can push the field forward, and the symposium is a test of that potential, not a validation of past credentials.
Consider what the question assumes. Counting first-author publications or weighing the reputation of a letter writer treats the invitation like a formula. If that formula worked, the selection process would be trivial, a script could parse CVs and generate invitations. The fact that ETH runs a symposium as the next stage tells you something important. They want to see how you think, how you engage with ideas you did not generate, and how you collaborate under pressure. A publication list tells them you can execute a project from start to finish. A symposium tells them whether you can do that inside a room full of people who might challenge your assumptions. Those are different skills.
What does this mean for you? If you are preparing an application, stop optimizing for the wrong signals. Focus on the depth of your work. A single well-crafted paper that opens a new line of inquiry is more compelling than five incremental ones. Choose recommenders who can speak to your intellectual curiosity and your ability to recover from a dead end, not just your productivity. And when you get to the symposium, if you get there, treat it as what it is: a conversation. The people on the other side of the table are not counting your citations. They are trying to decide if they want to work with you for the next three years. That decision depends on whether you can listen, adapt, and articulate a vision that is yours.
Stop chasing the checklist. Earn the invitation by being the kind of researcher who makes the checklist irrelevant.