Scott Reed is an AI research leader and entrepreneur with 11 years of experience bridging deep learning research and applied systems. After a long research tenure at DeepMind and a Principal Research Scientist role at NVIDIA focused on generalist embodied agents, he co-founded a new AI venture based in the Atlanta area. His academic foundation—a PhD in Computer Science from the University of Michigan—underpins work on unsupervised feature learning, generative models, and scalable object detection from collaborations with Google Brain. Scott has hands-on contributions to influential projects such as text-to-image GANs (ICML 2016 codebase), demonstrating both experimental rigor and practical engineering chops. He combines deep theoretical understanding with production-minded experimentation, often touching model architecture, data pipelines, and training systems. Colleagues describe him as a generalist researcher who moves fluidly between novel algorithm design and building the infrastructure to prove it at scale.
Contributions:11 commits, 9 pushes, 2 comments in 4 months
Contributions summary:Scott made several updates to various Lua scripts related to the generative adversarial text-to-image synthesis project. These updates include modifications to the main training scripts (main_cls.lua, main_cls_int.lua, main_txt_coco.lua) and demo scripts (txt2img_demo.lua), suggesting involvement in the training process and potentially model evaluation. The changes touch upon aspects such as model architecture, data loading, and parameter settings, implying an active role in the model development and experimentation.
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.