Builds virtual assistant (NLP), fraud detection, and predictive analytical systems in big-data production environments.A PhD social scientist with 20 years of experience with data munging, optimizing pipelines, deploying custom models, clear visualization & presentations, small team leadership.Machine Learning: Natural Language Processing with Rasa, spaCy & word2vec supervised deep learning models ensemble; classification & anomaly detection with regression, SVM & RandomForest trees supervised models; feature engineering & parameter tuning.Statistics: logistic & multi-variate regression; hypothesis testing; data cleaning & imputation; times series; variance & residuals analysis. Programming: R/RStudio with dbplyr, ggplot2, foreach, caret & shiny; Python/Jupyter with NumPy, Pandas, Matplotlib, Scikit-Learn & TensorFlow; HIVE & Postgre SQL in AWS S3 datalake; SQLServer; bash & beeline; Git/GitHub; AGM; Jenkins; Jira; Tableau.
Listed skills include R Shiny, Rasa Nlu, Scikit Learn, Consulting, and 66 others.