I currently lead the generative AI machine learning team at Scale AI. At Scale, I have led research on agents, alignment [1, 2], and reasoning and put AI services into production for data engine quality & efficiency. I created AFM-1, Scale’s first vision foundation model and led ML for human-AI collaboration in vision and language data engines. Our team developed & productionized models for spam, cheating, and fraud detection, LLM-based graders, and in-house custom reward models for quality control. The team I built drove business value out of AI research while simultaneously publishing our findings and acquiring top research talent.
I am currently focused on researching and building systems that most effectively leverage human supervision for the advancement of frontier models. Since 2015, I have been interested in developing models that can rapidly improve, by training powerful base models either though meta-learning, joint training, online learning, and improved calibration [1, 2]. Generally, I’m interested in technologies that can self-improve, collaborate with people, and improve human well-being.
Previously, I researched and developed deep learning systems at Standard Cognition, where I worked on video action recognition research, model training automation, transfer learning, domain adaptation, metrics development, hard mining, and model robustness. I also led migration of our core human pose estimation stack from tensorflow 1 to pytorch and implemented the real time production video inference service with TorchScript, Rust, and GStreamer. Before that, I was the first engineer at Explorer AI, an autonomous vehicle mapping company which was acquired by Standard Cognition.
I double mastered at the University of Arizona, focusing on machine learning and remote sensing, respectively. During that time, I was a researcher with the ML4AI lab in the School of Information. I was advised by Dr. Clayton Morrison and Dr. Greg Barron-Gafford.
X ~ personal github ~ linkedin ~ Scale AI github
Publications
Presentations
- Bringing Foundation Models to Automotive Data Engines. Tech.AD Europe 2024
- A Baseline Analysis of Reward Models’ Ability To Accurately Analyze Foundation Models Under Distribution Shift. AAAI Workshop on Responsible Language Models (ReLM 2024).
- Federated Reconnaissance: A New Framework for Distributed Continual Learning. Autonomy and Edge Computing session at the Fifth Annual Workshop on Naval Applications of Machine Learning (NAML 2021).
- Balancing Machine Learning Innovation and Scalability in Fast Growing Startups. Guest lecture in the Winter 2021 graduate course Machine Learning: Special Topics taught by Dr. Paul J. Atzberger at University of California Santa Barbara.
- Meta-Learning Initializations for Image Segmentation. NeurIPS 2020 Meta-Learning Workshop. https://meta-learn.github.io/2020/
- Continual Progress in Cascading Model Systems by Ensuring Reproducible State. Keynote speech at Deep Learning World 2020.