Alexandros Papakonstantinou

Designing compilers for enabling portable performance of AI models on heterogeneous systems at scale.

Technical Lead in the P4 compiler team for Tofino high-speed networking devices.

Lead in the architecture team of Samsung's mobile GPU.

Designed ML and HPC features for 5 generations of Nvidia GPU Streaming Multiprocessors (SMs).

Part-time employment at the University of Illinois, Urbana-Champaign during graduate studies toward the PhD. Research on: - High performance computing - Multicore architectures for parallel applications - FPGAs & GPU & CPU heterogeneous computing - High Level Synthesis and Compiler tool design - System on Chip & Embedded systems - Interconnection architectures, study of Network on Chip schemes

Graduate Intern Technical in Microarchitecture Research Lab, Intel Labs. Research on fast exploration of alternative multicore architectures through automated hardware design based on high-level language transformations and high-level synthesis techniques.

- Mobile x86 Processor Design Verification

- Floating Point Unit study and design - Tricore 2 RTL Design - Tricore 2 Integer Pipeline owner - Tricore 2 LD/ST Pipeline Designer - Synthesis Flow owner - Place & Route Flow contributor