Sana is interested in using compiler optimizations to help programmers of parallel machines exploit hardware features provided by modern GPUs so that they can automatically obtain the best performance for their applications. During her past internships with NVIDIA Research, she has worked on improving the performance of Monte Carlo and Ray Tracing applications on GPUs using program analyses and transformations, and also explored micro-architectural enhancements that address the problem of high control divergence in these domains.
Sana is a fourth year PhD student in Computer Science at Georgia Institute of Technology, where she is advised by Professor Vivek Sarkar. In addition to her research related to compiler optimizations for GPUs, she has also served as a teaching assistant for the GPU Architecture course taught by Professor Sudhakar Yalamanchili (ECE 8823) and as a co-instructor with Professor Sarkar for the course on Parallelizing Compilers (CS 6245) at Georgia Tech. Sana has previously worked as a software engineer at NVIDIA, India and has interned at Intel and NVIDIA Research.