• Building a service to predict milestones and improve on-time-performance. Analyzing data sources and running experiments with gradient boosting algorithms and neural networks. Building data pipelines and machine learning models in production.

• Built classification models to predict quote acceptance and estimate price elasticity. Used results as parts of revenue maximization models to generate optimal prices. Used regularization for feature selection.

• Built an engine with data scientists to automate and optimize shipment consolidation. The estimated ROI is a 3,100 hour cost-to-serve reduction per year. Wrote data pipelines, an optimization model, tests, and algorithms for a mixed-integer linear program.

• Built the workflow management and tracking system for setting prices with data scientists. Wrote the first models in production to optimize allocation management, with potential net revenue impact of $5-7 million per year.

• Founded, designed, and helped data engineers build the system for cost estimation covering all ocean and air spot contracts. It automated ingestion and loading of rate data, using a Gmail scraper and APIs. It was featured on the Flexport Engineering blog.

• Conducted research with Flexport's Chief Economist to present to executives. Built regression models to predict client profitability and ocean freight prices. Performed analyses on the trade war, which were featured in Flexport's blog and economic reports and cited on Bloomberg.

• Wrote scripts to extract and transform unstructured rate data from non-standard contracts, reducing processing time from one day to 20 seconds. Coded the first pricing models, reducing time to implement pricing and fulfillment strategy from six hours to one hour.