Mitchell Newman

Background

About Mitchell Newman

When I'm not two hundred cells deep in a Jupyter notebook, you might find me blowing off steam on the pickleball court or investigating the science of human personality and cognition. Technical Skills-Languages: Python, Golang, SQL, C++, Matlab -ML & AI: NumPy, Anaconda, Sci-kit Learn, PyTorch, h20.ai, Scipy, Classification, Regression, Clustering, Deep Learning, Feature Engineering, PCA and Dimensionality Reduction, Multi-layer Modeling -Stats: Descriptive and Inferential Stats, Hypothesis Testing, Confidence Intervals, Time Series -Practical Skills: Jupyter, Pandas, MySQL, PostgreSQL, Microsoft Office, Web Scraping, ETL processes, Faker, Data Visualization, Image Processing, Linux, log4j -Cloud: Microsoft Azure -Corporate Infrastructure: ServiceNow, Confluence, Jira, HashiCorp, Securiti.ai, Splunk Experience-Cybersecurity Data Science at Wells Fargo bank -Data Science Projects at Springboard with collaboration by senior professionals -Machine learning and cloud computing official accreditation from Microsoft, Coursera, and Udacity Some achievements (more work is featured in my GitHub profile)-Managed over 20 steps per month of scheduling, execution and validation of upgrades and vulnerability scans for Securiti.ai Kubernetes Pods, using ServiceNow, Linux and automated Machine Learning, ensuring seamless deployment of over 20 pods with new features added and validated on each monthly release -Implemented suite of unit tests for a custom log4j scanner, using Python, complex RegEx, and nested file archives such as ZIP and JAR, verifying the reliability of over 50 functions in the scanner -Built the highest accuracy random forest model for predicting official competitive tier for each Pokémon, inventing an original feature engineering method in dimensionality reduction based on Natural Language Processing -Created professional dataset for Kaggle using Python, Scrapy and extensive RegEx data cleaning of over 900 columns -Modeled the features and ticket price of over 200 US ski resorts to recommend pricing strategies and executive decisions, using Python, Pandas, Jupyter, Matplotlib and Sci-kit Learn, creating a strategy to increase Big Mountain Resort’s revenue by over $28 million for a season -Conducted hundreds of statistical tests using Python, Jupyter, Scipy and Distfit, finding outliers to make recommendations that helped improve accuracy of physiological signal reports for Cognitive Vultology by over 13% in one year Contact me at mitchellnewajdatascience@gmail.com if you want to chat about anything data science or machine learning related!