AI in the Real World with Foundation Capital

The promise of State Space Models: Karan Goel, Co-founder & CEO, Cartesia

Episode Notes

Welcome to AI in the Real World! In this episode, Foundation Capital Partner Jaya Gupta sits down with Karan Goel, co-founder and CEO of Cartesia, a company pioneering the use of state space models (SSMs) for AI applications.

Karan shared his journey from pursuing a PhD at Stanford to co-founding Cartesia with Albert Gu - taking their academic research on SSMs and turning it into a product that is initially focused on voice applications.

Karan explained how SSMs differ fundamentally from transformers, offering a more efficient, memory-based architecture that processes information sequentially rather than in batches. This approach enables more human-like AI systems that can remember and adapt to new information in real-time.

The conversation also covers:

• How SSMs offer significant efficiency advantages for applications like voice agents and on-device AI
• Why architecture innovation remains crucial alongside advances in data and training techniques
• The challenges and opportunities in the voice AI space
• His perspective on the open-weights vs. closed-model debate in AI
• Advice for researchers who want to become founders

Chapters:

• 00:05 Cold open
• 01:47 Karan's founder journey
• 03:45 Transitioning from academia to entrepreneurship
• 05:17 The art of recruiting
• 09:22 Karan's vision for Cartesia
• 13:36 Innovation in AI architectures
• 16:33 Advantages of State Space Models (SSMs)
• 19:34 The false dichotomy: models vs. applications
• 21:17 Best practices for voice AI
• 25:10 Thoughts on the broader AI landscape
• 31:37 Open source vs. closed models
• 35:00 Karan's thoughts on fundraising
• 41:27 Advice for founders building AI startups
• 46:03 Upcoming trends that get Karan excited