I am a PhD Student in the school of Computer Science at Georgia Institute of Technology. I am advised by Prof. Hyesoon Kim and was also priviledged to have worked under Prof. Sudakar Yalamanchili. My research interests are in the area of hardware accelerators and software co-design, focusing in the architecture design, programming languages and compiler tools to support heterogenous computing. I’m also interested in domain-specific applications of accelerators such as 3D graphics, graphs analytics, and machine learning.
I have a strong passion for teaching and doing research that promotes the accessibility and democratization of scientific knowlegde for the masses, contributing in several open-source projects and initiatives. The Vortex project is a “dream come true” initiative to open-up the full graphics processing unit software and hardware stacks, and provide a detailed microarchitecture pipeline that enables accelerators research in graphics, relational databases, graph analytics, and machine learning.
Prior to pursing my graduate eduction, I was fortunate to have had a long industry experience working on compilers, language, and simulation tools for graphics processors at Microsoft, playing at the intersection between the application developer, the operating system, and hardware. I also had the privilege to intern at many prestigious reseach groups including Catapult Project at Microsoft Reseach, Hardware Accelerator Research Program (HARP) at Intel Labs, Storage Systems Group at IBM Alamaden Research, Future Technologies group at Oak Ridge Labs, High-Performance Computing group at Pacific Northwest Labs.
Systems, compilers and hardware support for heterogeneous architectures.
Reconfigurable and hybrid architectures for high performance computing using FPGAs.
Software-hardware codesign of custom accelerators in Graph Analytics, Relational Databases and Machine Learning.
Novel architectures for low-power graphics accelerator in AR and VR domains.
High-bandwidth memory architectures and applications in high performance computing.