Shinya Takamaeda
Shinya Takamaeda researches computing systems for machine learning. His publications appear under the name Shinya Takamaeda-Yamazaki. His documented work includes the Pyverilog hardware-design toolkit, the QUEST neural inference engine, the STATICA annealing processor and newer work on memory-based computing and FPGA acceleration. Many of his papers report work conducted outside RIKEN. As of October 2026, Takamaeda directs RIKEN’s AI Computing Team, which studies devices, architectures and algorithms for efficient AI.
- Now at
- RIKEN
Timeline
-
2015
A tool for hardware design
Under the publication name Shinya Takamaeda-Yamazaki, he presented Pyverilog. The open tool processes Verilog hardware descriptions with a parser, data-flow and control-flow analysis, and code generation. It gives developers reusable building blocks for hardware-design software.
-
2018
Neural inference hardware
He coauthored QUEST, a neural-network inference engine reported at ISSCC 2018. QUEST uses a log-quantized design and stacked SRAM. He coauthored the paper outside RIKEN before his later team role.
-
2020–2025
Specialized computing systems
He coauthored work on the STATICA annealing processor in 2020, compute-in-memory designs in 2024 and an FPGA decision-forest accelerator in 2025. These projects address specific computations through specialized architectures.
-
As of October 2026
AI computing at RIKEN
Takamaeda directs RIKEN’s AI Computing Team. The team studies energy-efficient machine-learning systems across circuits, devices, architectures and algorithms. Its topics include hardware-aware learning and low-latency systems for serving large language models.
Sources and credits
Photo: www.riken.jp; Shinya Takamaeda profile photograph; creator not stated. License: Not stated by source. Source: riken.jp. Cropped to a square and resized. Ask us to remove or replace this photo.