ML Compiler Engineer (Member of Technical Staff)

• Developer on rocMLIR, MLIR-based open source compilation stack for AMD GPUs.
• GPU optimization features such as: KV-Cache, share memory swizzle for output stores, input fusions.
• New features such as: GEMM+elementwise+GEMM fusion (single kernel), shared memory automatic reuse.
• Support for new MI series architecture.

Senior Research Engineer

• Developed a fast panoptic segmentation model for Xavier NX achieving 30 fps.
• Contributed to an ICRA paper on deep learning-based depth completion.

Deep Learning Research Scientist

• Developed a fast panoptic segmentation model for Xavier NX achieving 30 fps.
• Contributed to an ICRA paper on deep learning-based depth completion.

Research Engineer in Computer Vision

• Authored a CVPR paper on deep learning-based multi-camera color constancy.

Research Assistant

• PhD thesis: Embedded 3D Reconstruction for Autonomous Driving.
• Adapt to GPU and parallelize (CUDA) computer vision algorithms.
• Computer Vision contributions: Developed a faster and more accurate version of the Stixel World.

PhD Internship

PhD Internship

• Developed a faster and more accurate version of the Stixel World (Computer Vision).

PhD in Computer Vision

Master's degree: Computer Vision

Bachelor's degree: Computer Science