Jack Chih-Hsu Lin

About Jack Chih-Hsu Lin

I am a lead data scientist at C3.ai. I have hands-on experience in ML (7 years) and DL (5 years) using Python (7 years). I build end-to-end DS/ML products for internal projects and external customers. I define problems and metrics; I devise/prototype solutions and productionize code. As a senior role, I lead the junior data scientists and communicate with software engineers, project managers, product managers, and UI designers. I also write weekly reports to executives and explain DS approaches to customers and stakeholders. From a PhD program to a tech company, I have developed predictive models for various types of data including tabular data, images, text, graphs, and have ranked as top 3.0%(26/866), 4.9%(188/3,835) and 6.1%(201/3,274) in 3 Kaggle machine learning worldwide competitions (overall rank: top 3.4%, 6,571/194,359). I also have invented, implemented, and published a new and interpretable neural network algorithm that converges 35% faster, reduces 200 times of parameters, and performs similarly to (AUROC>0.88) traditional neural network. With skills in data science, machine learning, and deep learning, I cannot wait to solve all these interesting real-world problems.• Languages: Proficient in Python (7 yrs), familiar with MySQL, Shell• Statistical analysis and hypothesis testing (NumPy, SciPy)• NLP: NER (Stanford NER, Hugging Face Transformer BERT, spaCy), fuzzy matching, TF-IDF, Word2vec• ML (7 yrs): regression, classification, clustering, random forest, Scikit-learn, gradient boosting (XGBoost, LightGBM)• DL (5 yrs): ANN, interpretable neural networks, image classification, object detection (Faster R-CNN), CNN (EfficientNet, EfficientNetV2), NLP (Siamese BERT), PyTorch, Keras• Graph machine learning: node classification, link prediction, random walk, PageRank, DeepWalk, node2vec, graph neural network (GraphSAGE, Graph Attention Network, R-GCN), Deep Graph Library (DGL)• Distributed computing (batch jobs; MapReduce), Docker, knowledge of Cassandra & Postgres, Pip/Conda, GitKaggle: https://github.com/Jack-Lin-DS-AIMedium: https://medium.com/@jacklindsai• Optimize PyTorch Performance for Speed and Memory Efficiency (2022)(10k+ views in a week): https://towardsdatascience.com/optimize-pytorch-performance-for-speed-and-memory-efficiency-2022-84f453916ea6• Self-Supervised Learning (SSL) Overview : https://towardsdatascience.com/self-supervised-learning-ssl-overview-8a7f24740e40

Experience

1. Software Engineer

   Google

   Mar 2025 — Present · Sunnyvale, CA, US

   GenAI for Workspace Gmail

Education

• National Taiwan University

  Master of Science (MS), Plant Pathology and Micriobiology

• National Taiwan University

  Bachelor's degree, Plant Pathology and Microbiology

• Baylor College of Medicine

  Doctor of Philosophy (PhD), Quantitative Computational Biology

  2020