ahmedmuaadh

Analyzed unstructured employee feedback using sentiment analysis within Amazon Fulfillment Centers to identify attrition risks and operational pain points. Presented data-driven solutions to Human Resources Management that would reduce burnout and turnover rates without negatively affecting business operations or profit.

Completed a simulation focused on quantitative research methods  Analyzed a book of loans to estimate a customer's probability of default Used dynamic programming to convert FICO scores into categorical data to predict defaults

Developed a modular AI pipeline leveraging OCR and PDF parsing to process over 200 pages of mortgage data blobs and streamline mortgage processing by automating data quality checks. Designed fixes for common OCR error patterns and implemented data quality checks to improve extraction accuracy to 96% and reduce the need for manual data review or correction. Built an interactive dashboard visualizing model KPIs, OCR outputs, and RAG context windows to monitor model performance over time and alert any performance degradations.

Designed and implemented a data collection workflow to collect product information, consumer demand, and style preferences to automate trend analysis and reduce workload for market analysts. Fixed issues present in other trend analysis models by correcting faulty API integrations and wrote documentation to facilitate the standardization of API calls across all other models. Consolidated model outputs and incorporated competitive benchmarks to improve readability of trend analyses and overall decision speeds.