MJ Llorente has 10 years of experience as a practicing data scientist with a focus on ecological modeling and ecosystem regeneration. As lead data scientist at Balance, MJ collaborates with the firm’s hydrologists to develop climate-aware models for evaluating restoration efforts in California watersheds, train predictive models on flooding events at Bay Area gauging sites, and consolidate decades of historical data for easier analysis. He also provides internal support for data reporting and has developed data modeling projects with colleagues and clients. Through these efforts, MJ continues his goal of applying scientific knowledge and methodologies to confront climate change. MJ earned his doctoral degree in materials science and engineering from UC San Diego and his bachelor of science degree in engineering physics from UC Berkeley, where he focused on carbon capture and conversion methods with renewable energy. MJ entered the realm of data science in 2017, becoming an instructor at Galvanize Data Science Immersive in 2018 and a founding instructor at Galvanize’s Los Angeles campus in 2020.

  • B.S. Engineering Physics, University of California, Berkeley, 2007
  • Ph.D. Materials Science & Engineering, University of California, San Diego, 2016
    • Perazzo Meadows and Lacey Meadows Machine Learning Studies. MJ evaluated the predictive strength of publicly available climate data and Balance streamflow data for late-season water retention and streamflow in the larger Lacey Meadows and Perazzo Meadows system. MJ continues to develop methodologies that use climate data models to estimate improvement in retention and flow from future restoration efforts in these systems.
    • San Mateo County Flood Warning System. MJ created time-series machine learning models and deep learning models with data from San Mateo County gauging sites to predict flooding events with maximal warning lead time.
    • Data Archive. MJ is organizing company site data, building a database infrastructure for internal company usage, and building data resilience and accessibility for both internal use and for clients with particular data needs.

    Few things are quite as fulfilling as witnessing the moment an audience member grasps a complex concept or realizes the importance of a result for themselves.

    Lake Merced, San Francisco, California