About

I'm Sara, an ML engineer interested in understanding how AI systems behave in the real world. I build evaluation frameworks, investigate model failures, and look for ways to make models more reliable without sacrificing the performance that made them useful in the first place. I earned my Doctor of Engineering in Artificial Intelligence and Machine Learning from George Washington University, where my research explored ways to reduce bias in image captioning models without retraining them.

Outside of work you'll usually find me in the garden. Lately I'm growing cut flowers and loofahs (yes, the sponge, they start life as an edible gourd!). I like watching things closely, experimenting, and learning from what actually happens rather than what I expected to happen. The mindset isn't all that different.