Assistant Engineer

YIZE XIE

*****@**.*** 617-***-**** Boston, MA linkedin.com/in/yzxie

EDUCATION

Boston University Boston, MA Expected May 2021

M.S. in Computer Science

• GPA: 4.00/4.00

Peking University Beijing, China July 2017

B.S. in Computer Science

• Major GPA: 3.70/4.00

Awards:

• Meritorious Winner for Mathematical Contest in Modeling (MCM)

• First Prize in Jiang Zehan Cup Mathematical Contest in Modeling (top 2/50)

• Third Prize in ACM Programming Contest by Peking University WORK EXPERIENCE

BU Artificial Intelligence Research Boston, MA Jun 2020 – Aug 2020 Intern as Research Assistant

• Designed multi-layer attention model for deep learning solution of fashion compatibility task (computer vision), wrote 2000 lines PyTorch code, achieved 3% better performance.

• Used python to clean, store and normalize data, practiced Object Oriented Programming in neural network. Computer Vision Research at BU Boston, MA Sep 2019 – May 2020 Research Assistant

• Implemented ResNet simulated neural network for transfer learning using TensorFlow with biological medical data, achieved 3 times better performance compared to paper, committed code regularly to GitHub.

• Debugged and refined codes from different version of PyTorch and to latest version to conduct experiments. Finup Co., Ltd. Beijing, China Dec 2017 – Dec 2018 Software Engineer (Natural Language Processing)

• Fetched 2TB data from Hive system every week, built Spark based word embedding system.

• Developed financial risk control system, deployed and visualized machine learning models such as K-means, LDA and Word2Vec, used ROC curve to explain and evaluate models, improved AUC from 0.6 to 0.75.

• Devised general solution for unsupervised feature generation system in risk control task, expanded original model with CNN (TextCnn), and formulated method to general dataset. PROJECTS

Airbnb House Price Prediction Boston University Sep 2019 – Dec 2019

• Ranked 1/83 in the Kaggle contest, Cleaned the noisy price data and developed SVM model (without libraries like sklearn), winning both public and private leaderboards. (all algorithms implemented in C++ for efficiency) Tic-Tac-Toe AI Contest Boston University Sep 2019 – Dec 2019

• Directed AI team with 2 other teammates, investigated papers, created optimized alpha-beta search algorithm and implemented chess playing robot with Java and OOP, ranked 1/30 in competitive tournament. SKILLS

Programming: proficient in Python, C/C++, MATLAB, Java, HTML, SQL, OOP Tools: proficient in Git, MySQL, Hive, JSON, OpenCV, OpenGL, Linux, Latex, Spark

Contact this candidate