Skip to content

Professional Experience

Professional

2025-present: Principal Research Scientist, Georgia Institute of Technology, Atlanta, GA, USA

  • Leading externally funded research projects as co-principal investigator, key personnel, and team lead
  • Serving on conference organizing/program committees, journal editorial boards, and specification committees in high-performance computing and programming models

2019-2025: Senior Research Scientist, Georgia Institute of Technology, Atlanta, GA, USA

Led and conducted research on:

  • An actor-based programming model and its runtime system for large-scale irregular applications
  • Enabling resiliency in asynchronous many-task (AMT) programming models
  • Multi-level programming models for Graphics Processing Units (GPUs) and SIMD units
  • Code generation and optimizations for GPUs
  • Automatic construction of compiler and runtime heuristics using machine-learning techniques
  • Programming models for quantum-classical systems

Additional responsibilities:

  • Led externally funded research projects such as IARPA AGILE as co-principal investigator, key personnel, and team lead
  • Served on conference organizing/program committees, journal editorial boards, and specification committees in high-performance computing and programming models

2015-2019: Research Scientist, Rice University, Houston, TX, USA

Led and conducted research on:

  • LLVM-based optimizations for large-scale systems
  • Code generation and optimizations for GPUs
  • Automatic construction of compiler/runtime heuristics with machine learning
  • Resiliency in AMT programming models

2013-2015: Postdoctoral Researcher, Rice University, Houston, TX, USA

Conducted research on compiler and runtime support for parallel languages such as Habanero-Java, focusing on high-level language constructs and parallel GPU code generation while maintaining precise exception semantics.

2012-2013: Assistant Professor, Waseda University, Tokyo, Japan

Conducted research on automatic parallelizing compilers and compiler-directed power reduction techniques in collaboration with:

  • TOYOTA and Denso (automotive engine control software)
  • Mitsubishi Electric (dose calculation software for heavy-ion therapy)
  • Fujitsu Ltd. (smartphone applications on Android)
  • Olympus Corporation (multimedia applications)

Taught freshman-level courses in computer literacy and C programming (Spring/Fall 2012).

2010-2012: Research Associate, Waseda University, Tokyo, Japan

Conducted research on automatic parallelization and power reduction techniques for embedded heterogeneous multi-core processors. Most notably, optimized a clinically-used dose calculation program in collaboration with Mitsubishi Electric, achieving over 50x speedup on IBM POWER7 platforms.

2007-2010: Research Associate, Global COE Program, Waseda University, Tokyo, Japan

Conducted research on automatic parallelization and power reduction for embedded heterogeneous multi-core processors in collaboration with Hitachi Ltd. and Renesas Electronics Corporation. This work was supported by the Global COE program “International Research and Education Center for Ambient SoC,” funded by Japan's Ministry of Education, Culture, Sports, Science, and Technology (MEXT).

Teaching

Courses

  • Computer Literacy, Waseda University, Spring 2012. (English Course)
  • C Programming, Waseda Univerisity, Fall 2012. (English Course)
  • Science and Engineering Lab., Waseda Univerisity, Spring/Fall 2012. (Japanese Course)

Tutorials

  • Fundamentals of CUDA Programming, Waseda University, Spring/Fall 2018
  • Fundamentals of FPGA Programming, Waseda University, Spring 2018

Awards

  • Best SCALE Challenge Award, IEEE TCSC International Scalable Computing Challenge at the 24th IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGRID2024), May, 2024. (Youssef Elmougy, Akihiro Hayashi, Vivek Sarkar) LINK
  • Outstanding Research Scientist Award, College of Computing, Georgia Institute of Technology, 2024 LINK
  • Best SCALE Challenge Award, IEEE TCSC International Scalable Computing Challenge at the 23rd IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGRID2023), May, 2023. (Youssef Elmougy, Akihiro Hayashi, Vivek Sarkar) LINK
  • IEEE Senior Member, 2021
  • Outstanding Paper award, The 2020 International Conference on High Performance Computing & Simulation (HPCS2020), Mar 2021. (Tiago Carneiro, Nouredine Melab, Akihiro Hayashi, Vivek Sarkar) LINK
  • Best Expo Exhibit, 28th Annual International Conference on Computer Science and Software Engineering (CASCON2018), November 2018 (1 best exhibit out of 72 exhibits). (Akihiro Hayashi, Gita Koblents, Max Grossman, Kazuaki Ishizaki, Alon Shalev Housfater, Jimmy Kwa, Vivek Sarkar) LINK (A PDF File will open)
  • Best Expo Exhibit, 27th Annual International Conference on Computer Science and Software Engineering (CASCON2017), November 2017 (2 best exhibit out of 54 exhibits). (Akihiro Hayashi, Gita Koblents, Max Grossman, Kazuaki Ishizaki, Alon Shalev Housfater, Jimmy Kwa, Vivek Sarkar) LINK (A PDF File will open)
  • Encouragement award, IPSJ Symposium on Embedded Systems, October 2013. (Dan Umeda, Yohei Kanehagi, Hiroki Mikami, Akihiro Hayashi, Mitsuhiro Tani, Hiroshi Mori,Keiji Kimura, Hironori Kasahara)
  • Best feature award, COOL Chips XVI, IEEE Symposium on Low Power and High-Speed Chips, April 2013. (Yohei Kishimoto, Hiroki Mikami, Keiichi Nakano, Akihiro Hayashi, Keiji Kimura, Hironori Kasahara,) LINK
  • Encouragement award, IPSJ Symposium on Embedded Systems, October 2012. (Yohei Kishimoto, Hiroki Mikami, Keiichi Nakano, Akihiro Hayashi, Keiji Kimura, Hironori Kasahara)
  • IPSJ SIG recommended Ph.D thesis, September 2012.
  • Best presentation award, 5th GCOE Symposium, September 2009.

Selected Grants

  • IARPA AGILE: Flow-Optimized Reconfigurable Zones of Acceleration (FORZA)”. PI: Vivek Sarkar, co-PIs: Tom Conte, Alex Daglis, Akihiro Hayashi, Hyesoon Kim, Rich Vuduc (Georgia Tech), Scott Beamer (UC Santa Cruz), Peter Kogge (U. Notre Dame), Dean Chester, Gunnar Gunnarsson, Todd Inglett, Brian Smith (Cornelis Networks), Marty Deneroff, Tim Dysart (Lucata), Dave Donofrio, John Leidel (Tactical Computing Labs). Total Award Amount: $15,847,862 (shared among six institutions).
  • ORNL STAQC: “Software Stack and Algorithm for Automating Quantum-Classical Computing”. PI: Jeffrey Young, co-PIs: Akihiro Hayashi, Tom Conte, Vivek Sarkar Total Award Amount: $358,000