Skills

The surface of a corroded magnesium sample.
Image taken with Fisher Scientific Nova 200 Nanolab

Focused Ion Beam cross section image of corroded magnesium sample. In order from top to bottom: platinum deposit, needle structure, corroded layer, magnesium matrix.
Image taken with Fisher Scientific Nova 200 Nanolab

A cross section of a corroded magnesium sample. Domes of corrosion product can form over particles (seen in the center) during aqueous corrosion
Image taken with Fisher Scientific Nova 200 Nanolab

A closer look at the particle in the center of the dome
Image taken with Fisher Scientific Nova 200 Nanolab

Magnesium surface after aqueous corrosion. This sample is "uniformly" corroded though domes still form. A closer look would reveal the needle like structure of the magnesium hydroxide.
Image taken with Fisher Scientific Nova 200 Nanolab

Raw data from a Statsmodels dataset. This forecasting was done following the Guide to Time Series Forecasting article by Digital Ocean. Though this example is done directly following a guide the same principles and methods were applied to much of my work at United States Steel. The ARIMA (Auto Regressive Integrated Moving Average) method is a statistical method that can be applied to time series data sets for forecasting. 

Various plots generated by statsmodels in Python to indicate the quality of predictor model.
Top left: Residuals - mean of zero, no apparent seasonality
Top right: KDE and N(0, 1) follow closely, indicating a normal distribution
Bottom left: Blue dots follow linear trend, another indication of a normal distribution
Bottom right: Correlogram shows little correlation between a value and its lagged versions. Indicates little to no seasonality in residuals.

Testing the ARIMA model on existing data to ensure accuracy

Predicting 40 years (480 months) into the future. The light shaded pink indicates the confidence interval.