Omar ElTayeby

Data Scientist

Lowe's•  June 2020 - Present

• Worked with a team of 4 Data Scientists to identify data quality problems with Lowe’s e-commerce website then proposed and implemented new solutions that improved our vendors’ landing process • Built an outlier framework to detect errors improving the accuracy to 78% (+40% improvement) • Applied classification algorithms on endless aisle to categorize the products with 85% F1-score • Built a control selection framework as an alternative for A/B testing based on cosine similarity of products’ KPIs • Implemented these algorithms and frameworks on Google Cloud Platform (i.e., Big Query, Airflow)

Graduate Assistant (RA/TA)

University of North Carolina at Charlotte•  August 2014 - May 2020

• (RA) Social Media User Annotation: visualization tool to iteratively categorize Twitter users • (RA) Personalized Curiosity Engine: a personalized recommendation system for a cooking app • (RA) Learning Analytics: a project for assisting faculty leaders in reducing the risk of student attrition • (RA) Event Detection: detect and predict planned events across the globe in real-time via Twitter • (RA) Demographic Visualization: a visual analytics system that enables the exploration of Reddit users’ interests • (TA) Database Systems: Graded assignments and mentored students for their group projects • (TA) Visual Analytics: Gave tutorials on building dashboards using Tableau, D3, and R Shiny

Informatics Specialist Intern

Mayo Clinic Rochester•  May 2017 - August 2017

• Aim: Develop a new Public Health Index to optimize resource allocation for disease research • Analyzed over 3.5 MM archived news articles from Reuters to reveal their coverage of medical conditions • Applied NLP techniques to identify changes in the publics’ perception of diseases over time • Found that some viruses and unhealthy habits were associated with unrelated infectious diseases

Graduate Assistant (RA/TA)

Clark Atlanta University•  August 2012 - May 2014

• (RA) Media Influence analysis: a project for measuring the media’s influence on Twitter users • (RA) Cancer genes analysis: Measured the interaction stability of a DNA-protein complex in a cancerous cell • (TA) Software Engineering: Prepared material and lectured software development life cycles to graduate students

User Assistance Intern

Oak Ridge National Laboratory•  May 2013 - August 2013

• Aim: Develop an interface to monitor the usage of a High-Performance File System (Lustre) • Built a time-series visualization to monitor the storage and input/output usage • Compared the performances of two JavaScript libraries (D3 and Highcharts) for building the interface