Question about timing of program: This program is during the school year rather than during summer break. It is also meant for EU/UK students, who may not have slack during the school year because EU/UK university admissions often specifically require very high grades with little room for error. Do you think your application pool would be stronger if this were a summer program instead? 

(pointed) Questions about "puzzle quiz": 

1. These synonym questions (and to a lesser extent the analogy questions) are dramatically easier for native English speakers. What about EU students who are non-native English speakers?
   
   
2. The test displays your results at the end (below is an excerpt of my results from clicking randomly). If you give students a number, that's going to give them much more reason to Goodhart the shit out of it, speculate about cutoffs, etc, etc. Also a person might simply feel inadequate looking at the results of what is essentially an IQ test heavy on English vocab and probability theory. Do you have a reason for displaying these results? 
   

Sounds like a great idea. Just want to mention that I love the design!

Exciting project! I'm curious to see the results of the fellowship. Also wondering if you think the program model is replicable to target students outside of the US/UK? I think with some contextualization, it has a lot of potential to be impactful at scale.