Model quality, hallucinations, and how LLMs work

Subscribe to AI Agents Podcast Channel on YouTube: https://link.jotform.com/subscribe-to-podcast

In this episode of the AI Agents Podcast, host Demetri Panici sits down with David Petrou, founder and CEO of Continua AI, to discuss the evolution of AI agents, multimodal models, and social AI.

They explore how agentic systems are reshaping collaboration, productivity, and human connection, sharing insights from real-world experience building AI products and lessons learned from past innovations.

David shares his journey from Google, where he spent nearly two decades working on AI and machine learning, to founding Continua AI, a company focused on creating AI agents that actively participate in group conversations, nudging people to connect and collaborate more effectively.

They discuss the nuances of AI development, from early experiences with machine learning and language models to the latest breakthroughs in reasoning, tool usage, and zero-shot learning.

David and Demetri also cover technical insights, including AI hallucinations, token efficiency, latency, cost optimization, and on-device processing, providing listeners with a clear understanding of the opportunities and challenges in building intelligent systems today.

This episode is a must-watch for anyone interested in the future of AI, productivity, and social intelligence—whether you're a developer, entrepreneur, or AI enthusiast—and how cutting-edge AI can elevate both personal and organizational performance.

▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬

⏰ TIMESTAMPS:

00:00 – Proactive AI agents and Slack integrations

03:18 – Introduction to the AI Agents Podcast

05:15 – David Petri’s background and journey into AI

09:37 – Machine learning, humility, and agentic intelligence

14:06 – Why early AI ideas failed and lessons from Google Glass

18:36 – Multimodal AI and real-world use cases

23:03 – Hallucinations, tokens, and how LLMs actually work

27:03 – Cost, latency, and the rise of on-device AI

31:20 – Designing social AI for group conversations

36:48 – Product vision: AI that enhances human connection

41:55 – The future of agentic AI and collaboration

47:32 – Final thoughts on innovation and timing

52:11 – Episode wrap-up and closing remarks

▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬

Sign up for free ➡️ https://www.jotform.com/

▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬

Follow us on:

Twitter ➡️ https://x.com/aiagentspodcast

Instagram ➡️ https://www.instagram.com/aiagentspodcast

TikTok ➡️ https://www.tiktok.com/@aiagentspodcast

▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬