Ep.13 Jeff Chou - Sync Computing скачать в хорошем качестве

Ep.13 Jeff Chou - Sync Computing

In this episode of Unapologetically Technical, I interview Jeff Chow, CEO and co-founder of Sync Computing. Jeff, who holds a PhD from UC Berkeley and a postdoc from MIT, shares his unique journey from academia to startup life, and how his experience with simulations shaped the vision for Sync Computing. This episode explores: The challenges of translating academic research into real-world solutions. The crucial need for bridging the gap between technical execution and business objectives. A fascinating look at analog distributed systems and their potential for solving complex problems. An in-depth explanation of Spark cluster cost optimization and the impact of factors like skew, spot instances, and data size. Sync Computing's approach to declarative computing and its potential to simplify resource management and cost optimization. Valuable advice for both aspiring entrepreneurs and those considering a career in academia. This episode offers a rich discussion for data engineers, software engineers, and anyone interested in the intersection of technology and business strategy. 0:00:00 Introduction 0:01:25 Importance of Simulations 0:08:11 UC Berkeley 0:11:09 GenAI in Teaching 0:14:35 Writing Optical Simulations 0:26:01 Analog and Quantum Computing 0:34:47 Getting Close to the Business at UPS 0:40:59 What Is Sync Computing 0:48:42 How Sync Works 0:55:24 Dealing With Skew 1:04:47 What Contributes to Spark Cluster Cost 1:08:15 Manual vs Automatic Cluster Optimization 1:15:39 Clouds Supported 1:22:07 Changing Parameters Between Runs 1:26:00 Convincing To Leave or Join Academia 1:31:58 Difficulty of Entrepreneurship 1:35:42 Passion and Reasons to Do Startup 1:38:04 Ways of Thinking About What's Important (20 Years)