• Leading development of custom global supply chain network design optimization for footwear, apparel, and accessoriesusing Python (pandas, Google OR Tools) by transitioning from Supply Chain Guru X thus saving $1M in licenses fees• Enhancing ML models in Python (pandas, scikit-learn) that forecast production and tariff costs for new products for aproduct sourcing optimization model that generates recommendations based on factory utilization and cost• Maintaining ML models in Python (pandas, scikit-learn) that generate estimates for footwear product material and laborcosts, the results of which are used by 150+ designers in a dashboard built using Snowflake and Tableau• Acting as scrum master by leading daily stand ups, sprint retrospectives, sprint planning sessions, sprint demos, and software architecture design sessions while tracking and organizing team progress using Jira, Confluence, and Excel• Championing the resourcing of new team members and teaching software development and data science best practices

• Leading data mining and analysis for the development of a munitions sourcing and distribution optimization model and dashboard application on the Palantir Foundry platform using Python (pandas, pyspark, PuLP)• Developing an encoder-decoder neural network model to predict backorders for 30,000 items within a 12-month time series forecast horizon using TensorFlow, achieving an averaged mean absolute error (MAE) that ranges from 0.75 to 1.0• Building Multilayer Perceptron and Recurrent Neural Network (RNN) models for aircraft engine component reliability prediction using unstructured text for 30+ classes using scikit-learn and TensorFlow achieving an accuracy of 85%• Sourcing, cleansing, and transforming data from Oracle and PostgreSQL databases using SQL, pandas, and pyspark• Applying SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-Agnostic Explanations) to explain model predictions and generating performance metrics using scikit-learn and leveraging matplotlib to visualize results• Leading weekly Foundry application development sessions with external teams and mentoring team members on the usageof the Foundry platform and data science techniques• Creating data imputation pipeline to fill missing item characteristics based on nomenclature similarity using fuzzy string matching using fuzzywuzzy and external data sourced via web scraping using scrapy

• Designing and implementing temporal U-Net neural network models to generate segment maps of solar surface phenomena using satellite images, resulting in an increase in accuracy of 106% over the previous model in Keras with a TensorFlow backend using 2 GTX 1080 Ti GPUs• Building end-to-end pipeline to preprocess over 30,000 observations (300 GB) collected by NASA’s Solar Dynamics Observatory, correcting image luminosity, and augmenting images in Python (scikit-learn, NumPy, and pandas)• Performing literature review of state-of-the-art convolutional neural network (CNN) and recurrent neural network (RNN) computer vision models, tuning hyper parameters, producing briefings with visualizations generated in matplotlib, and collaborating with senior data scientists on Gitlab

• Developed and implemented dual input recursive neural network models to classify hate speech in Wikipedia and Twitter comments via natural language processing, TensorFlow, and Keras using an RTX 2070 GPU and increased performance by 0.65% over baseline models• Generated sentence embeddings using the Google Universal Sentence Encoder and bias via Word Embedding Factual Association Test (WEFAT) association scores from GloVe word embeddings using Python (NumPy, pandas, and scikit-learn)• Performed literature review on state-of-the-art hate speech detection methods and produced technical report to document methods and results

• Developed enhancements for a predictive item grouping recommendation engine in SAS and SQL that identified optimal item groups for long-term contacts using item characteristics and Federal Acquisition Regulation (FAR) sourcing requirements improving client’s workload prioritization• Led prototyping of a supply chain network dashboard in Qlik Sense that monitored distribution center performance metrics using Python and SQL• Performed ad-hoc analyses on the effectiveness of historical item grouping strategies for prioritizing client workloads, item churn rate, and overall performance of recommended groups by querying Oracle and SAP HANA data warehouses• Coordinated timely internal invoice review and tracking for project work orders totaling over $6 million in annual revenue

• Coordinated with project leadership to monitor SDLC phases (schedule variance, scope and quality) and produced PowerPoint presentations used for communicating the status to stakeholders on a weekly basis• Supported project management and release management teams to produce financial deliverables for client review and sign-off and assisted with the creation and forecasting of new scope• Facilitated weekly meetings with team leads for risk/issue management, implementation plan, and financial/staffing reviews and maintained compliance for a team consisting of over 100 resources• Developed Python and VBA (Excel) scripts to automate manual processes via the generation of email content and financial reports saving approximately 5 man hours per month and ensuring accuracy of reports

• Supported the engineering and industrial design teams by developing medical safety device designs using SolidWorks and manufacturing rapidly prototyped models with 3D printers• Executed tension and compression testing on prototypes using a motorized test stand, acquired experimental data, and performed statistical analysis on the results to determine structural and material enhancements for future designs• Evaluated the structural performance of 3D printed and machined models through experimentation, engineering theory and finite element analysis in order to determine failure points under applied loads• Designed​ ​3D printed fixtures and developed testing documentation for pilot-scale product testing