About

Chenzhao (Chris) Li is an experienced data scientist with a strong background in statistical modeling, machine learning, and deep learning. With a PhD in Civil Engineering (Statistics track) from Vanderbilt University, he has authored over 20 publications and received 200+ citations, as well as two best student paper awards. As a Machine Learning Engineer at Facebook, Chris specializes in autoencoder, computer vision, CNN, and Bayesian Network. He has over 12 years of relevant experience, having worked as a Senior Data Scientist at Panton Inc. and a Staff Data Scientist at Baker Hughes. At Baker Hughes, Chris developed an autoencoder model from scratch for 10+ subsystems of 3+ years time series data and contributed to automatic hyper-parameter tuning using Bayesian method. He also applied MC Dropout algorithm to calculate real-time anomaly probability and researched on LSTM, Variational Autoencoder, cyclic learning rate, and other related topics. Additionally, he built a CNN to detect 10+ classes of symbols from 200+ piping and instrumentation diagrams (P&ID) and improved mAP from 0.67 to 0.75 by anchor box optimization and data augmentation. At Panton Inc., Chris developed multiple models using transfer learning and CNN to classify roof penetrations, automatically detect hail damages on roof, and classify house siding images. He also integrated regression and clustering models to segment 3-D point cloud for a drone-enabled roof measurement system and reduced the system failure rate from 20% to 2% by tuning hyper-parameters. Chris is proficient in GCP and UI and has a strong tech stack that includes ML, CNN, object detection, data scientist, research scientist, object detection, computer vision, and deep learning. He is not in need of sponsorship and is a valuable addition to any team seeking a highly skilled and experienced machine learning engineer.