Prashant Garg
I have always found it hard to start from one narrow question. Since PhD application season, I have cared less about choosing the first thing I could supply quickly than about finding the question where an hour of attention might matter most.
Early in my PhD that made focus difficult. Later, working across papers and teams, I developed a specialization in measurement with machine-learning tools and network ideas while keeping fairly eclectic topic interests. FrontierGraph came out of that tension. As AI makes it easier to produce more work, the direction problem matters more, not less.
FrontierGraph maps open economics questions that sit near existing work. It helps narrow what to read next, but it is not a substitute for judgment.