Scientist

Junkai Zeng ******@**.*** 540-***-****

Education Ph.D. in Physics (Quantum Computing), Virginia Tech, 2014 - 2019 (expected) Graduate Certificate in Data Analytics, Virginia Tech, 2018 B.S. in Physics, University of Science and Technology of China, 2010 - 2014 Skills Programming Languages: Python, C++, Matlab, Mathematica, SQL, Unix Shell (Bash) Technical Tools: Numpy, Scipy, Pandas, Scikit-Learn, Tensorflow, Matplotlib, Seaborn, Jupyter, Google Cloud Platform (GCP), Git, HTML, MySQL, MongoDB Exceptional research, quantitative, and problem-solving skills, include statistical modeling, machine learning, data visualization, numerical analysis and simulation; Strong knowledge in data structures, algorithms, and object-oriented programming. Experience Research Assistant, Virginia Tech, Blacksburg, VA, 2015 - Present

• Proposed a new theoretical protocol on noise-resistant quantum control for quantum computing devices;

• Independently discovered a geometrical structure hidden within the Schrödinger equation that connects quantum mechanics and differential geometry;

• Developed a verification framework in python to test the quantum control fidelity;

• Presented in academic conferences and collaborated globally with peer scientists;

• Published 3 research papers as 1st author.

Research Intern, Institute of Quantum Computing, Waterloo, Canada, 2013

• Developed, documented, and maintained Matlab program to simulate the physical process of electron and nuclear double resonance;

• Communicated with the experimental team to modify the theoretical model based on real experimental data.

Projects Instacart Market Basket Analysis

• Exploratory data analysis on history transaction record to extract statistics on each user and product;

• Built a machine learning model to predict whether an item will be ordered by the same user, based on xgboost classifier. Achieved a result of 0.382 in the metric of F1-score. NYC Taxi Trip Duration Prediction

• Cleaned, explored, and visualized over 1 million taxi trip records in Jupyter;

• Divided NYC map into different areas by using K-Means and DBSCAN clustering algorithms on taxi pick-up and drop-off location data;

• Extracted statistics on trip duration in different scenarios;

• Predictive modeling based on gradient boosting regression. Deep Reinforcement Learning for Multi-Agent Soccer

• Investigated and implemented Deep Reinforcement Opponent Network(DRON), an expanded version of deep Q-network, for modeling the primal agent, the team agents, and opponents policy simultaneously on a multiagent task of 2D soccer game. Activities GSA Delegate, Elected physics representitative at the Virginia Tech Graduate Student Assembly, 2015 - 2017 Candidate, Cheers Science Season 3 Episode 1 & 2, a Chinese reality television show about science broadcast on CCTV-1, 2018

Contact this candidate