Darryl Barnhart
About Darryl Barnhart
Darryl Barnhart is the Cofounder and CTO of Latent Space, where he manages large GPU clusters and develops advanced data pipelines for generative models. He has a diverse background in software engineering, having worked at notable companies such as Google and TD Bank Group, and holds multiple degrees in computer science and engineering.
Current Role at Latent Space
Darryl Barnhart serves as the Cofounder and Chief Technology Officer (CTO) at Latent Space, a position he has held since 2019. In this role, he is responsible for managing large GPU clusters, which are essential for the company's operations in generative modeling. His expertise in technology and leadership contributes to the development of innovative solutions within the organization.
Previous Experience in Technology
Before joining Latent Space, Darryl Barnhart accumulated significant experience in the technology sector. He worked at Google as a Software Engineer from 2013 to 2017, where he contributed to various projects in Mountain View, California. His earlier roles include positions at Angle Technologies and TD Bank Group, where he served as a Software Engineer and IT Analyst, respectively. Additionally, he worked as a C++ Programmer at Health Canada in 2005.
Educational Background
Darryl Barnhart has a diverse educational background in computer science and engineering. He studied at the University of Waterloo, earning a Bachelor of Science in Computer Engineering from 2004 to 2006. He furthered his education at Ontario Tech University, obtaining a Bachelor of Science in Physical Science from 2010 to 2013. He also completed an Advanced Diploma with Honours in Game Programming at Humber College from 2007 to 2010. In 2013, he participated in Summer Intensive Studies in Computer Science at Stanford University.
Technical Contributions and Projects
Darryl Barnhart has made significant contributions to the field of technology, particularly in the area of generative models. He has been involved in the design and implementation of modeling frameworks and data pipelines for applications in language modeling and computer vision. His work includes developing high-performance hybrid GPU/CPU vector indexing for datasets exceeding 1 billion scale, utilizing differentiable stochastic sampling techniques.
Research Experience
In 2012, Darryl Barnhart worked as a Research Assistant at the University of Toronto for four months. During this time, he was affiliated with the Bahen Centre for Information Technology, where he gained valuable experience in research methodologies and technology applications.