🎓 PhD Student, Computer Science — NC State University
🧑‍🏫 Supervised by Dr. Tim Menzies

🧠 Software Engineering × Machine Learning × Optimization

📉 Optimization in Software Engineering
Why do many SE optimization problems collapse to small search spaces — and how can
simple, sample-efficient optimizers beat heavyweight methods?

🤖 ML for Software Engineering (ML4SE)
Data-driven models for software decisions under noise, uncertainty, and limited labels.

🧪 Testing & Fuzzing (as instrumentation)
Using fuzzing/testing to generate signals (failures, behaviors, causes) that feed
optimization + learning—especially for systems-level evaluation.

🧩 Many SE problems are over-modeled
📦 Data is expensive; labels even more so
⚖️ Complexity should be earned
🔍 If a method can’t explain itself, it can’t be trusted

📌 Sample-efficient multi-objective optimization
📌 Search-space collapse (BINGO effect)
📌 Lightweight optimizers for noisy SE tasks
📌 Data-centric evaluation via testing/fuzzing

💻 Python Java C/C++ JavaScript Bash
📊 ML · optimization · empirical SE