Amit Sabne

I am a software engineer at Google Brain. I work on high performance compilers for Tensor Processing Units (TPUs), named XLA.

Before, I was a software engineer at Microsoft, working in the Visual C++ compiler team. I researched and designed novel optimization techniques to improve program performance while lowering binary size and compilation time.

Earlier, I earned a PhD in Computer Engineering from the School of Electrical and Computer Engineering, Purdue. I worked with Rudolf Eigenmann in the Paramount Lab. My PhD research area was High Performance Computing, with a focus on heterogeneous computing systems. My dissertation proposed and developed efficient programming models for accelerators. It also provided compiler and runtime support for these models and formulated fast autotuning mechanisms for accelerator programs.

While in grad school, I did several internships. I interned with the Integrated Systems Group of NEC Labs in the summer of 2015. During Fall 2013, I worked as an ASTRO program researcher at Future Technologies Group, Oak Ridge National Labs. I spent the summers of 2013 and 2012 interning at nVIDIA in the CUDA Compiler team. In the summer of 2011, I interned with the Modems Systems Engineering team at Qualcomm.

Before joining Purdue, I worked at BHEL, Industrial Systems Group (ISG) in the Drives and Automation Lab. Even before that, I received a B. Eng. (Hons.) in Electronics and Instrumentation from Birla Institute of Technology and Science (BITS)-Pilani, India. I interned with Texas Instruments and Bharat Forge during this time.