- Developed and maintained deep learning models for computer vision applications- Proficient in MLOps practices, with hands-on experience using various MLOps-related tools and frameworks to streamline the machine learning workflow.- Collaborated with cross-functional teams to gather requirements, implement models, and deploy them in production environments.- Conducted thorough evaluations and fine-tuning of models to achieve optimal results, considering factors such as performance, efficiency, and scalability.- Contributed to research and development initiatives, exploring innovative approaches to solve challenging computer vision problems.- Stayed up-to-date with the latest advancements in computer vision and deep learning, continuously exploring new techniques and methodologies to enhance model performance.- Mentored junior team members, providing guidance and support in their learning and professional growth.

- Managed databases, performed dataset preprocessing, and applied labeling techniques using tools.- Conducted evaluations of cutting-edge methods, structures, and frameworks on GPU platforms, utilizing languages such as C++, Python, and CUDA.- Developed and trained AI models for various computer vision tasks, including image classification, object detection, semantic segmentation, and image-to-image translation.- Provided optimized models and production-level sorting solutions, ensuring high-quality and efficient performance.- Collaborated in CI/CD processes, assisting with the deployment of pre-trained and fine-tuned models, utilizing tools.

- Developed prototypes and generated technical reports using tools such as Matlab, Python, and R.- Collaborated effectively with overseas teams, utilizing version control systems like svn and git, as well as project management tools like Jira.- Optimized compute-intensive applications, specifically focusing on 1D and 2D convolutions, as well as memory-intensive tasks.- Implemented designs on acceleration hardware, leveraging languages such as C++ and OpenCL for FPGA, OpenCL and CUDA for GPU, and OpenCL and OpenMP for multicore processors.

Assisted lectures and labs of the following courses: --ECE Induction Lab--COMPSYS202 (C/C++)--SOFTENG206 (Java)

Developed three FPGA-based PXI cards (Xilinx Virtex-5 and Zynq SoC), responsible for• Schematic design and circuit debugging• Developed IP core (Verilog), BSP, and embedded applications (C/C++)