Yousef Eddin

Using adversarial noise to defeat Automatic License Plate Readers.

• Problem: Flock Safety is tracking people and selling their data to ICE whether or not you're suspected of a crime.
• Approach: Build a permissioned dataset and benchmark ALPR detection/OCR robustness under controlled synthetic perturbations; catalog failure modes to guide safer system design and privacy-focused mitigations.
• Outcome: Don't know yet!

Analyzed revenue by supplier, lead times, and the tie between delivery speed and financial performance.

• Problem: Identify top suppliers and bottlenecks across the chain.
• Approach: Exploratory data analysis with SQL/Python; feature pivots and visualizations.
• Insight: Clear pattern between on-time delivery and higher revenue share.

Built a front-end tool to search TFT team compositions by level and by units I already have, ranking results to minimize decision time.

• Problem: Quickly identify the best comp to pivot into based on current board state.
• Approach: Level-based filtering with a “What do you already have?” selector; comps scored by unit overlap and sorted by closeness-to-complete.
• Outcome: Highlights the most achievable comps first while clearly surfacing missing units and active bronze traits.

Used Python to extract, clean, and apply machine learning to predict housing prices.

• Problem: Estimate home prices based on complex market features.
• Approach: Data cleaning, feature engineering, and XGBoost regression model.
• Outcome: Achieved strong predictive accuracy and feature insights for key housing factors.

Used Python to clean job market data and build regression models to predict salaries.

• Problem: People are often under paid. How do I know how much I should be paid?
• Approach: Cleaned a large dataset, then used Linear Regression to predict how much salary should be.
• Insight: Strong features in experience level and company size; produced interpretable salary trends.