Build Real-Time Voice Agents: Architecture Walkthrough

I have spent 15 years making the world's most complex tech at Google and Microsoft easy to understand through visual storytelling, and now I help companies do that through my business of advising and consulting.

Along the way, I built and shipped AI products, including launching Gemini Code Assist at Google and leading developer go-to-market for GitHub Copilot at Microsoft.

I teach AI for Business at Wharton, guide Fortune 500 CEOs through AI transformation, and was invited to speak at TED on the human skills that matter most as AI takes over. My two books, Visualizing Google Cloud and Visualizing Generative AI, are bestsellers, have 350K+ developer community.

My belief is simple: you shouldn't need a computer science degree to understand and lead technology.

Kris Efland is VP of Engineering at Deepgram, where he leads teams turning cutting-edge speech AI into practical tools for developers. Over the past two decades, he has built consumer products, autonomous systems and global cloud services. Before Deepgram, Kris led Amazon Personalize and AWS’s Geospatial AI and machine learning portfolio, including the launch of satellite-imagery capabilities for Amazon SageMaker. Earlier, he led autonomous mapping and AI tooling at Lyft and held senior engineering roles at HERE and Nokia. His work focuses on building AI systems that customers can use reliably in production.