Costa is Senior VP and Chief Scientist at Mad Street Den Labs. He is well-known internationally as an experimental and computational neuroscientist in the area of single neuron electrophysiology. He has been the principal investigator on grants from the National Institutes of Health and others to study how neurons encode and transmit information. After taking the leap from academia to industry, he built perhaps the first large-scale GPU-based neuron simulator, long before GPU's were widely used in machine learning. Subsequently, he developed advanced imaging and classification systems for the Naval Research Labs, NASA, and DHS before coming full circle to improve neuronal simulation with GPU's for NIH.
He holds MD and PhD degrees in neuroscience from the University of Virginia along with biomedical engineering, electrical engineering and computer science degrees from Johns Hopkins University. He leads MSD's charter alongside Anand, building future neural network architectures that can enable more generalizable models of intelligence.