•Built custom prompt flows and model configurations in Azure AI studio to deploy a RAG model on company application data, providing more structure and accessibility for data lineage and data governance tasks•Utilized SQL for maintenance of company data quality rules resulting in a 17% increase in usable data

•Created lessons and assignments in R programming and statistics for a class of 20 students•Facilitated one on one tutoring in college level probability courses

•Utilized Twitter API, SQL, and Pandas to collect and process a dataset of 100,000 tweets concerning the term ‘gender ideology’ in Brazil from 2013 to 2023•Created an LDA model for topic analysis of Brazilian tweets, leading to 15 different unique topics where discussion of ‘gender ideology’ in Brazil is present•Applied Transformers libraries to fine-tune a pre-trained BERT model, resulting in 82% accuracy when analyzing sentiment of ‘gender ideology’ tweets•Created an interactive… Show more •Utilized Twitter API, SQL, and Pandas to collect and process a dataset of 100,000 tweets concerning the term ‘gender ideology’ in Brazil from 2013 to 2023•Created an LDA model for topic analysis of Brazilian tweets, leading to 15 different unique topics where discussion of ‘gender ideology’ in Brazil is present•Applied Transformers libraries to fine-tune a pre-trained BERT model, resulting in 82% accuracy when analyzing sentiment of ‘gender ideology’ tweets•Created an interactive data timeline in Python that synced activity on Twitter directly with Brazilian political events that occurred from 2013 to 2023•Generated insights into organized Twitter campaigns by Brazilian politicians and recurrent trends on social media centered around election cycles Show less

•Leveraged SQL & Oracle Database to wrangle and clean a large dataset of auto claims from 2019 to 2021 for model training•Analyzed dataset in R and built a predictive model for cost using generalized linear models•Examined auto claim costs over time and created data storytelling visualizations in Excel that provided critical business insights regarding increased accident severity from 2019-2021

•Designed a data annotation pipeline using AWS Sagemaker for Ph.D research on hate speech detection •Collaborated with Ph.D. students to develop annotation guidelines for hate speech detection on Twitter in English and Spanish resulting in a more efficient hate speech detection algorithm across languages•Reviewed and provided comments on crowd/vendor annotated data in English and Spanish to identify violations in guidelines•Utilized Pandas to process Twitter datasets for training in… Show more •Designed a data annotation pipeline using AWS Sagemaker for Ph.D research on hate speech detection •Collaborated with Ph.D. students to develop annotation guidelines for hate speech detection on Twitter in English and Spanish resulting in a more efficient hate speech detection algorithm across languages•Reviewed and provided comments on crowd/vendor annotated data in English and Spanish to identify violations in guidelines•Utilized Pandas to process Twitter datasets for training in hate speech detection algorithm Show less