Matt Schmill, PhD

Data Science


Machine Learning and Artificial Intelligence
Natural Language Processing
Metacognition and Metareasoning
Software Engineering


Ph.D. Computer Science, University of Massachusetts, Amherst
M.S. Computer Science, University of Massachusetts, Amherst
B.S. Computer Science, University of Massachusetts, Amherst



Recent Projects

International Lifestyle Clothing and Accessories Retailer
Voice of the Customer (VoC) Analytics
Developed a suite of models to mine customer opinions, feedback, and preferences across ratings, surveys, social media text and imagery, and call center data. Modeled features to tie behavior to business outcomes. Segmented customers across several dimensions, including tendencies, demographics, and products. Created models for real-time scoring in GCP.

Multinational Construction Company
Predicting Fiber Line Routes in Urban Construction Environments
Inferred the relational structure of utility poles and fiber spans from complex data without relational structure. Developed algorithms and models to predict routing of fiber lines in field settings. Integrated models in SME workflows using Airflow and Spark.

Technical Expertise

  • Python data stack: scikit-learn, pandas, numpy, Jupyter, others
  • Supervised and Unsupervised Machine Learning,  Probabilistic models, Temporal models and time series analysis, Word Embeddings, BERT, Vector space models, Named Entity Extraction
  • Hadoop: Spark, MapReduce, Hive, Airflow, Luigi, Nifi
  • PostgreSQL, Oracle, SQL
  • GCP, BigQuery, Cloud Functions, EC2, RedShift, EMR, Lambda, S3