Cambrian AI Visions: Meet the Leaders
Karl Freund with Cambrian AI sits down with key newsmakers and influencers in the world of Artificial Intelligence.
Subscribe to our newsletter to get first notification when a video is posted.
Qualcomm Technologies’ Sr Director, Product Management John Kehrli discusses MLPerf benchmarks
Qualcomm released MLPerf benchmarks for image and language processing that surprised a lot of industry watchers. In this short video, Qualcomm Technologies' Sr...
Cambrian AI Presentation: Top AI HW Since Last AI HW Summit
As Presented by Karl Freund at the AI HW Summit Topic: New AI Hardware Technologies and Benchmarks Karl presented his top 10 list at the annual AI HW Summit again this...
Jeff Gehlhaar, Qualcomm Technologies VP of Technology and lead for Qualcomm AI Software.
Qualcomm Technologies surprised everyone with exceptional performance and efficiency in the MLPerf 1.0 benchmark release. Jeff Gehlhaar leads the AI software team at...
Andrew Feldman of Cerebras Systems, making 120 Trillion parameter models possible.
Our Second Interview With Andrew Feldman, Founder and CEO Cerebras Systems, discussing the need for massive models and the Cerebras technology that makes 120 Trillion...
Interview with Ziad Asghar, Qualcomm Technologies Inc. VP of Product Management
Qualcomm Technologies Ziad Asghar discusses the current and future AI in mobile and the data center.More Cambrian-AI Visions Video Interviews
Interview with Mike Vildibill, VP of Product Management of Qualcomm Technologies, Inc.
Qualcomm Technologies surprised everyone with exceptional performance and efficiency in the MLPerf 1.0 benchmark release. Mike Vildibill, GM of the Cloud AI100 business...
Interview with Mike Henry, Founder and CEO of Mythic
Mythic has just launched their first analog compute inference engine, with the idea of putting computation INSIDE flash memory. Yes, Flash! Also, see a recent Latest...
Interview with Luis Ceze, CEO of OctoML
OctoML provides a compilation service to ML developers using Apache TVM, allowing them to their models to virtually any CPU, GPU, FPGA, or AI accelerator....