This was one of the coolest classes I've ever taken, and Professor Agueros was the perfect fit. Because of the low number of people taking the course, he replaced the midterm and final this semester with a midterm presentation (on really cool astrophysics topics like gamma ray bursts and Type Ia supernovae) and an oral final (everyone studied the same paper for this final, which was about the effects of mass loss on the evolution of high-mass stars). Both were pretty enjoyable to study for and they were DEFINITELY preferable to taking a legit midterm or final. He did end up asking some difficult questions on the spot, but he'd always try, at least, to point you towards the answer.
The lectures were always interesting. I slept less in this class than I have in any other class, which says a lot. Professor Agueros had a lot of really neat Powerpoints, he'd always derive the equations (but wouldn't take a crazy long time with derivations), he always spent the right amount of time between the powerpoint and the board, and - this is, to me, the best thing - he ALWAYS made sure to emphasize exactly why what he was talking about was important. Always. It was really, really refreshing to experience a professor do this, especially in a class where some of the math can get away from the physical meaning. We'd always have to think about the results we got in a larger context (i.e. "What does this mean for the star?")
The problem sets were, to be quite honest, pretty tough. But at the same time, they were the kind of "tough" wherein you see the problem and know you have all the tools to solve it - it's just a matter of thinking through it and figuring it out. Weirdly, I think I would've enjoyed more homework in this class, because there were always really cool applications of concepts in class that you'd never think of. But at the same time, having only four PSets was a blessing.
Finally, Professor Agueros made a pretty conscious effort to introduce guest speakers, fringe topics, and scientific computing to the class. A lot of problem sets had a computational component, in which we had to compute some stellar models and answer questions about them. It was very different than what we were used to, but you could tell that he was trying to form the same habits that he and all the guest lecturers already had. The guest lecturers were always relevant to what we were discussing and they too were vulnerable to being asked difficult questions by Professor Agueros on the spot, which was (sadistically) kind of fun to watch.
Overall, this was a GREAT class - if you're really interested in astronomy and stars, I would definitely recommend you take it.