Registered user since Sat 18 Oct 2025
Sangeeta Chowdhary is a compiler and systems researcher working on GPU runtimes, scheduling systems, and compiler technologies for large-scale AI and accelerator-driven workloads. She is currently a researcher at AMD Research, where her work focuses on compilers, runtime systems, and efficient execution strategies for modern accelerators.
She earned her Ph.D. in Computer Science from Rutgers University, where she worked on debugging numerical errors, floating-point analysis, and compiler-based program analysis systems under the supervision of Prof. Santosh Nagarakatte. Her research has appeared in leading systems and programming languages venues including PLDI, ESEC/FSE, OOPSLA, and WAHC.
Her broader research interests span compilers, GPU systems, AI infrastructure, runtime scheduling, numerical analysis, and hardware/software co-design for machine learning systems. She has also contributed to research projects involving homomorphic encryption compilers, LLVM-based systems, and high-performance runtime optimization.
Before pursuing research, she worked in industry on networking and distributed systems at Samsung and Mahindra Comviva, designing high-performance networking and mobile systems.
Contributions