Anthony Ngo Email & Phone Number

Kirkland, WA, US

Bothell, Washington, United States

Communications Lead-Build Loss models- Flesh out data pipelines

Redwood City, California, United States

• Conduct analysis 1 million records of data to discover and report insights to other Data Science team members
• Utilize PySpark and SQL techniques to engineer data processing pipelines to prepare 1.2 million records of dataper month for modeling
• Develop ensemble classifiers using web activity to anticipate whether a user will chat with Customer Service
• Inspect feature importances to gather insights from weaker performing models and inform and implementmodel improvements to increase model performance by 4%
• Document project information including project purpose, methodology, takeaways, and next steps

Seattle, Washington, United States

• Develop bespoke data migration scripts to transfer data from 3rd party data sources into the Wildlife Insights platform, expanding the platform dataset by 1,050,000 images
• Design Python scripts to assist image migration and validation
• Validate that outputted datasets adhere to the Wildlife Insights batch upload template using Postman
• Document necessary packages and instructions to migrate data using the developed scripts
• Collaborate with 6 partner companies to optimize and execute data migration
• Organize the data engineering workflow for incoming developers, aiding the onboarding of 3 new hires

Seattle, Washington, United States

• Employ Sklearn to design classification models that classify cognitive impairment with accuracies of 70%-90% and ROC_AUC Scores of 0.75-0.95
• Conduct error analysis to evaluate the demographics in which my classification models fail
• Produce 9 regression models to predict the performance of patients on various neuropsychological exams to evaluate the alignment of speech and cognitive function
• Review academic literature to form a foundation for further research and gather speech-based features with strong predictive power

Greater Seattle Area

• Clean, process, and join 4 datasets from the Framingham Heart Study
• Spearhead exploratory data analysis regarding cognitive decline to discover trends and relationships between vocal markers and dementia using Python and Jupyter Notebook and natural language packages
• Present, explain and justify my findings to the lead research advisor in biweekly meetings
• Develop Elastic Net Regression models to predict cognitive performance with an accuracy of.73 using trends from my analysis
• Draft and design several data visualizations for fellow researchers
• Coauthor an academic paper on the vocal indicators of degenerative neuropsychological diseases alongside a team of 7 graduate and post-doctoral researchers

Master Of Science - Ms, Computer Science

Bachelor Of Science - Bs, Computer Science, 3.7