I am the Chief Scientist at Distributional, where we're building an automated testing platform for deep statistical testing of AI applications. I am also the founder of DataScientific, Inc., an AI Advisory and Consulting firm specializing in the development and implementation of cutting-edge AI solutions. Previously, I was the Chief Machine Learning Scientist at H2O.ai, a leading AI company known for producing H2O, an open source, distributed machine learning platform, along with Driverless AI, h2oGPT, LLMStudio, and a range of other Enterprise AI systems. My tenure at H2O.ai was marked by the creation and leadership of the development team for the H2O AutoML algorithm (the first open source enterprise AutoML platform), where I also spearheaded efforts in explainable/interpretable AI, algorithmic fairness and AI benchmarking and measurement.
Additionally, I am the founder of WiMLDS (Women in Machine Learning and Data Science) and a co-founder of R-Ladies Global, both organizations aimed at promoting diversity and inclusion in the AI field. I also collaborate with the OpenML organization to develop open source benchmarking tools for machine learning, including the industry standard benchmark for AutoML systems (AMLB).
Author or co-author:
- H2O: Scalable Machine Learning & AutoML Platform
- H2O AutoML Wave App: Wave App (web GUI) for H2O AutoML (Python)
- h2o4gpu: R interface for H2O4GPU, machine learning on GPUs
- rsparkling: R interface for H2O Sparkling Water, machine learning on Spark
- OpenML AutoML Benchmark (AMLB): Benchmarking Framework for AutoML tools (Python)
- cvAUC: Computationally efficient confidence intervals for CV AUC estimates in R
- subsemble: R package for ensemble learning on subsets of data
- SuperLearner: R package for Super Learning (Stacked Ensembles)
- meetupr: R interface to the meetup.com API
- rHeathDataGov: R interface to the HealthData.gov Data API