Summary
Oscar Skean is a research-focused software engineer and PhD student in Computer Science at the University of Kentucky, specializing in representation learning and information theory for self-supervised computer vision. With nine years of engineering experience, he bridges rigorous research and production-quality code, having trained CNNs and Transformers to state-of-the-art performance and developed practical estimators and information-theoretic losses that improve sample efficiency. He has applied his skills in industry through roles at Triton International and a machine learning engineering internship at Stripe, and is passionate about leveraging HPC to scale large ML workloads. Beyond model performance, he explores scalable techniques to quantify and explain the knowledge embedded in multimodal neural networks, combining theoretical insight with hands-on systems engineering.
9 years of coding experience
3 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at University of Kentucky