Database, API, and AI development

Developed a new method to quantify the uncertainty of predictions from random forest machine learning models for applications in business forecasting, predictive analytics, and AI explainability.

• Leveraged machine learning, mixed-effect and generalized regression models, and other parametric and nonparametric statistical techniques to improve operating efficiency for multiple Fortune 500 companies • Led data science workstreams of engineering analysis projects• Constructed a pipeline to automate marketing overhead with tuned large language models for natural language generation and classification tasks• Developed, optimized, and marketed 3 websites as a full-stack developer• Forecasted, modeled, & visualized data with Excel & VBA, PowerBI, and Tableau

• Crafted a machine learning model to predict course demand for 3000 courses and 35000 students• Surpassed previous prediction method by 300% in multi-step out-of-sample time series forecasting• Integrated Bayesian Markov Chain Monte Carlo models to quantify prediction credibility• Engineered and selected 100+ autoregressive features utilizing walk-forward validation and data visualization, presenting to technical and non-technical stakeholders • Developed 2 R packages and 1 Python package to integrate models seamlessly with existing architectures, designing end-to-end data pipeline