● Saved $1.5 Million annually in vendor costs by forecasting traffic with xgboost (Python, Airflow)● Reduced financial planning time by 90% and improved sales forecast accuracy by 2% by creating automated time series forecasts for using pytorch and tensorflow (Python, Airflow, SQL, AWS) ● Won 4 awards for building source-of-truth reporting: data mart with 12 tables (130K queries/year) and 6 dashboards (58K views/year), including the 2 most used finance dashboards (SQL, Tableau) ● Increased annual sales 5% by building a structural equation model (SEM) to optimize marketing and inventory for each stage of digital and store conversion funnel (R, SQL)● Identified and led cross-functional team to add location data to traffic, identity resolution, and transaction pipelines, increasing geographic resolution up to 8M times, saving $100K annually in vendor costs; these data support models that contribute $75M in annual sales● Identified gaps in marketing and event data and built tables to fill the gaps; these data support models that contribute $64M in annual sales (SQL)● Built trust by managing data mart automation/security and API data extraction across 4 database migrations with 5 vendors and 4 teams (finance, engineering, product, analytics)● Optimized queries and pipelines to reduce run time and cost by up to 94% (SQL, Python, Airflow)● Reduced data pipeline failure rate by 40% by setting up data quality alerts (JSON, Pagerduty, SQL)● Audited board and quarterly earnings reports to ensure accuracy and consistent definitions (SQL)● Developed onboarding documentation used 800 times and SQL training used by 100-person team

● Used machine learning to make staycationLA, a web app that eliminates the health risks, carbon footprint, and cost of international travel by recommending Los Angeles county attractions that are similar to international destinations ● Leveraged NLP pattern detection over large datasets (7 million travel records and 170,000 reviews from a travel review website) scraped and cleaned in Python using selenium and beautiful soup ● Optimized recommender algorithm using topic modeling and cosine similarity in scikit-learn and nltk, improving MAP@10 by 200% and coverage by 600% compared to popularity baseline● Engineered a front-end website with flask, HTML, and Bootstrap on AWS

● Applied statistical modeling and simulation to generate business insights for pharmaceutical development and conservation by assessing how humans choose plants to use as medicine; for example, identified demographic traits associated with knowledge of specific illnesses or plants● Tested 60 hypotheses using exponential random graph modeling, generalized linear mixed models, and structural equation models in R with data from 80,000 plant uses and 305 people● Predicted 34-89% of variation in people’s medicinal plant knowledge based on human social network structure, human demographic data, and plant accessibility using statistical modeling

● Generated insights to optimize conservation by assessing where climate change will have the largest negative impacts on plants used for medicine and weaving● Leveraged pattern detection over large datasets by querying, integrating, and cleaning integrating 10GB of geospatial data and 4,000 GPS points from 15 databases● Developed 7 machine learning algorithms in R to classify and map presence/absence of 2 plant species across Australia and New Zealand; increased model sensitivity and specificity to 77-88%● Forecasted machine learning models using predictive modeling; quantified model uncertainty and mapped change in species survival likelihood due to climate change ranging from 40% increase to 12% decrease

Taught 9 lecture and laboratory classes in botany and ecology; created curriculum, mentored 150 undergraduate students, and received teaching award nomination

Developed writing, presenting, and teaching skills to communicate science for the public and mass media

Designed and taught inquiry-based curriculum for high school students to characterize organisms and explain relationships with genetic and morphological data

● Designed and implemented experiments to test how climate change may affect late blight, a fungal pathogen that threatens potato and tomato crops around the world that causes millions of dollars of damage annually● Tested growth rates for 60 lineages of late blight from three populations at four different temperatures and with two fungicide concentrations ● Collaborated with geneticists to identify genetic regions associated with growth rate and fungicide resistance using genome-wide association to improve pathogen management with continued climate change and increasing fungicide resistance

Field and lab exploration of plant taxonomy, phytochemistry, & pharmacology. Conducted individual research project, “Biomedical Potential of Native and Invasive Orchids in the Dominican Ethnobotany”.