Ozgur Tasdizen
Principal Gpu Architect @Arm
Signup · Get unlimited contacts
WORK HISTORY
Principal Gpu Architect @Arm
Cambridge, GB
I define the hardware architecture of next-generation Mali GPUs with a focus on scheduling behavior, PPA-optimized microarchitecture, hardware–firmware interaction, and long-term architectural direction. I propose and evaluate new architectural ideas by modifying the GPU model, running benchmarks to analyze PPA, and providing recommendations to engineering teams. I occasionally develop RTL prototypes to validate concepts or unblock implementation. My work spans compute, fragment, and geometry pipelines as well as memory subsystem behavior across GPU generations.Prior to moving into the GPU team as a Principal GPU Architect in 2019, I worked across the ISP and Display teams, gaining broad system-level experience across multimedia IPs. As a Principal Research Engineer in the Display team (2018–2019), I contributed to Arm’s AR/VR display-processing roadmap, including lens distortion and chromatic aberration correction, frame-rate up-conversion, and foveated rendering. As a Principal Engineer in the ISP team (2016–2018), I translated concepts into implementable hardware architectures; developed a fixed-rate compression approach and supporting hardware microarchitecture targeting visually lossless compression at practical ratios; designed an AC lighting flicker reduction algorithm and hardware microarchitecture for rolling-shutter image sensors; and evaluated HLS toolchains to assess the feasibility of integrating them into the development workflow.
EDUCATION
Istanbul Technical University
B.Sc., Electronical and Telecommunications Engineering
Sabanci University
M.Sc., Microelectronics
Deutsche Schule Istanbul
High School
Sabanci University
Ph.D., Microelectronics
SKILLS
ABOUT OZGUR TASDIZEN
Principal GPU Architect defining the hardware architecture of next-generation Mali GPUs, with a focus on PPA-optimized microarchitecture and long-term roadmap direction. My earlier work across video, display, and ISP pipelines gives me a strong system-level understanding of multimedia workloads and their performance bottlenecks. I’m motivated by designing efficient, scalable GPU architectures for future mobile and low-power compute platforms.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.