Software, data, and strategy. Trained at Carnegie Mellon University in Qatar, shipping hybrid ML systems, database architectures, and the occasional open-source contribution to Apache Lucene.
↓ Scroll to work
Pittsburgh
I'm a senior at Carnegie Mellon University in Qatar, majoring in General Studies with a minor in Business Administration. My coursework leans into machine learning, databases, and software engineering on one side, and strategy, marketing, and operations on the other, and I try to keep one foot firmly in each world.
Recent work spans a Ridge regression model for dengue fever forecasting (selected from four candidates after time-aware cross-validation), a hybrid TF-IDF + BERT classifier over 500K+ Reddit posts for mental-health query answering, a contribution to Apache Lucene's merge scheduler (mentored by Amazon engineers), and client-facing consulting projects on pricing and positioning for the Tepper side of campus.
I spent a semester on exchange in Pittsburgh, took a Global Learning Trip to Morocco, and generally enjoy work that happens at the intersection of things: disciplines, languages, geographies.
2024 to 2025 · 7 selected
Summer contribution to Apache Lucene, mentored by Amazon engineers. Designed a shared merge scheduler that coordinates merges across multiple IndexWriters, with global executor sharing, task tracking, and graceful shutdown for multi-tenant workloads. Reviewed through GitHub PRs against Lucene's core architecture.
Practical Data Science (67-364) with Peidi Dong. Trained and compared four regression models on 18 years of weekly dengue data from San Juan and Iquitos. Engineered seasonal, lagged, and rolling features; used TimeSeriesSplit to avoid temporal leakage. Ridge selected as final model (RMSE 6.89, R² 0.67).
Trained logistic-regression, random-forest, and XGBoost classifiers on 500K+ Reddit posts using a hybrid TF-IDF + BERT embedding pipeline. Built a concern-level scoring system from rule-based logic and model confidence. Presented to faculty.
Designed inheritance relationships, foreign-key constraints, triggers, and functional-dependency analysis to BCNF. Wrote 10 Python user stories with JOINs, aggregations, and window functions.
Tepper course 70-437. Conducted behavioral analysis to identify priority shopper segments and their implications for growth. Refined scope, prioritized strategic questions against client constraints, and delivered recommendations the team could act on.
Detected user communities with the Louvain algorithm and evaluated modularity for cohesion. Quantified preference diversity across stratified clusters.
Course 70-381. Translated consumer-perception research into segmentation and positioning direction, and recommended messaging built around authenticity as the primary value driver.
Not a transcript, just the ones that keep showing up in my work. A mix of computer science, machine learning, and the Tepper and Dietrich classes that made the technical material land differently.
Helped establish the iX Lab, a research space now home to a growing body of IS faculty projects. Early-stage work focused on organizing the physical and operational setup so the lab could become what it is today.
Developed a SharedMergeScheduler to enable multi-tenant merge coordination across IndexWriters. Refactored task logic for global executor sharing and graceful shutdown handling, reviewed via GitHub PRs against Lucene's core architecture.
Two weeks across Casablanca, Marrakech, and Essaouira. Visits to five women's cooperatives, exposure to Amazigh craft and culture, and learning that ranged from the Atlas Mountains to the coast. The kind of trip that quietly changes how you read a room.
Part-time on-campus role. Led a full rebuild of the Information Systems department's physical inventory, from a disorganized state to a fully catalogued, photo-indexed, and labeled system. The result is an inventory the department now actually uses.
Cross-campus exchange from CMU-Q. Coursework across machine learning, marketing, and software engineering. A lot of walking Forbes Avenue.
Interdisciplinary track combining computer science and ML coursework (Imperative Computation, Machine Learning, Databases, Software Engineering, AI in Medicine, Practical Data Science) with Tepper business courses (Strategic Management, Marketing, Operations, Organizational Behavior).
A short write-up from the mental-health query project. Where classical features beat embeddings, where they didn't, and the hybrid scoring trick that ended up mattering most.
Read → Coming soon · OSSNotes from the summer I spent contributing to Lucene's merge scheduler. On design reviews, mentorship from Amazon engineers, and why the best part was the stuff that never got merged.
Read → Coming soon · StrategyNotes-to-self after the Strategic Management capstone. What structured problem-solving actually means, how to scope, and why "it depends" is usually cowardice.
Read →