Avani Popat

Key Responsibilities & Achievements:

Built a feature selection pipeline for multi-omics data including proteomics (60 samples, 620 features), lipidomics (60 samples, 213
features), and metabolomics (60 samples, 943 features) totaling 1,776 features across three patient groups.
• Used Random Forest and Gradient Boosting Machine to compute feature importance, selecting top 70 features overall and
v

• Supported ongoing data science and machine learning research initiatives within the university's Data Science department on
a part-time basis
• Assisted in collecting, cleaning, and preprocessing academic datasets to prepare them for downstream analysis and modeling
• Contributed to exploratory data analysis and early-stage model development under the guidance of research staff
• Helped document

• Produced reproducible Python/Jupyter Notebook projects demonstrating end-to-end ML workflows and core data science
concepts for learners.
• Developed interactive Power BI dashboards to visualize project outputs and explain feature importance and results to non-
technical stakeholders.
• Authored technical blog posts and LinkedIn educational content (concept explainers and interview guides) to su

• Designed and delivered an end-to-end SQL curriculum using SQL Workbench and NIIT LMS covering basic to advanced topics.
• Developed lesson recordings, exercises, and assessments for self-paced learners.
• Collected participant feedback via course surveys (self reported 95% satisfaction) and iterated on content based on learner
responses.

Pursued a Master's degree in Data Science with a focus on Artificial Intelligence and Machine Learning in Finance. Completed a thesis on feature selection for classification of omics data.