Built and deployed an end-to-end deep learning pipeline using Python, PyTorch, and OpenCV, covering data preprocessing, augmentation, and model evaluation.
Fine-tuned ResNet and EfficientNet architectures, achieving a 12% increase in classification accuracy and a 30% reduction in inference latency compared to baseline models, significantly improving model performance.
Applied model pruning and quantization techniques to optimize the model for edge deployment, ensuring faster, resource-efficient inference on edge devices.
Automated data pipelines for large datasets (e.g., prices, macro indicators, alternative data) using Python and SQL, streamlining data ingestion and processing. Implemented anomaly detection and validation checks, reducing processing time by 30%.
Machine Learning Intern
Shenzhen Research Institute of Big Data
Shenzhen, China
07.2020 - 09.2020
Collected and curated a large dataset of medical images (X-rays, MRIs, CT scans) with corresponding diagnostic labels to support machine learning model development.
Experimented with various model architectures (including CNNs and deep learning approaches) to identify the most effective one for medical image analysis.
Optimized models through hyperparameter tuning to enhance prediction accuracy, significantly improving the model’s robustness and efficiency.
Education
Master of Engineering - Computer Science
Cornell University
Ithaca, NY
05-2025
Bachelor of Engineering - Computer Science and Technology