About me

I’m a PhD student at the Department of Statistics and Data Science, Wharton School.

I am fascinated by the success of perturbation genomics, with CRISPR technology as the leading example. Realizing the full potential of CRISPR experiments requires statistical methods that address calibration, power, and computation jointly — weaknesses on any one axis undermine the value of the others. My research advances all three fronts:

  • Calibration and robustness. Perturbation-based testing (e.g., resampling, permutation, and subsampling) for sharp Type-I error control under low signal-to-noise ratio and complex null dependence. Representative works: pcit, symcrt.
  • Power. Compound power analysis along two tracks: practical platforms for experimental design, and theory through the compound decision framework and nonparametric techniques. Representative works: PerturbPlan.
  • Computation. Mathematical theory drawing on classical asymptotic analysis to keep methods scalable and provably reliable. Representative works: spaCRT.

For the principles guiding this work, see my Research Vision.

In addition to research, I’m passionate about teaching and mentoring since I am deeply grateful for and benefit from the supervisors I have met, who have taught me far beyond just statistics. I have been a mentor for Undergraduate Research in Probability and Statistics and Directed Reading Program at UPenn.