Show Notes
In this episode, I talk with Bas Brekelmans, CTO of Copilot Studio. The conversation covers the evolution of conversational AI models, challenges in managing and deploying the latest models, the impact of model evolution on performance and security, and the shift in building agents from specific workflows to evaluating expected outcomes. The discussion also delves into the history and evolution of conversational AI, early models and innovation, challenges of model management, model selection and orchestration, security and guardrails in model orchestration, performance and latency in model orchestration, and the evolution of building agents. The conversation delves into the impact of AI models on engineering discipline, the transition from traditional development to model training, and the evolution of skill sets in natural language understanding. It also explores the role of language proficiency in AI development and the evolution of AI systems and models.
Takeaways Evolution of conversational AI models Challenges in managing and deploying the latest models Impact of model evolution on performance and security Shift in building agents from specific workflows to evaluating expected outcomes Nudging AI models Transition from traditional development to model training
Chapters 00:00 Evolution of Building Agents 30:06 Automating Reordering and Over-Engineering 36:12 Evolution of Engineering Discipline 41:33 Transition to Managerial Role in AI Development 48:10 Evolution of AI Systems and Models
More from The Agent Dudecast
In this episode of the Agent Dude Cast, host Dewain Robinson interviews James Oleinik from Microsoft, discussing the evolution of business applications, the introduction of the 'super app' concept, and the significance of context in AI-driven solutions. They explore how Microsoft's innovations, particularly in Copilot and Work IQ, are transforming productivity and automation in enterprise environments. The conversation delves into the advancements in AI and continuous learning within business applications, particularly focusing on Work IQ and its implications for CRM systems. James Oleinik and Dewain Robinson discuss the importance of automating processes and the potential for AI to enhance efficiency and personalization in business operations.
Dewain introduces his career transition and new role at Dynamic Communities, followed by a discussion with Uli about their initial meeting and relationship building. They delve into career progression, mentorship, and the importance of governance in AI. The conversation then explores model optimization, routing, and the challenges of operating AI at scale. The conversation delves into the key aspects of scaling AI, focusing on identity, observability, and the role of MCP as an enablement and control point. It also explores the challenges of agent development, the future of SaaS and AI, and the role of SaaS in the AI ecosystem. Additionally, it discusses physical AI and governance concerns, highlighting the need for responsible and ethical AI practices.
The conversation covers a wide range of topics related to AI, agents, and the evolving landscape of technology. It delves into the impact of AI on various industries, the importance of reinvention, and the role of AI in amplifying human capabilities. The discussion also emphasizes the need for thought leadership and innovation in the context of AI and technology. The conversation covers a range of topics, including the impact of AI on business, the concept of human context protocol (HCP), the importance of side hustles, and the evolving role of AI in consumer products and branding. It also delves into the distinction between AI in corporate and personal spaces, and the need for businesses to maintain a human touch in their interactions.
Podcast Summary The episode introduces the evolution of the Agent Dude channel, shifting to practical, real‑world use of agents and Copilot. It covers the role of the Copilot Acceleration Team (CAT) and CAPE in helping organizations adopt, scale, and govern AI. Key themes include the Frontier Transformation Pattern for AI maturity, the Power Up program for accelerating learning, and the Guidance Center and live webinars as core enablement resources. Boot camps, hands‑on labs, and the Copilot Studio Kit are highlighted as accelerators for building governed, scalable agents. The conversation closes by encouraging individuals and organizations to work with AI as a collaborator to amplify productivity, creativity, and impact. Chapters 00:00 – Introduction and Channel Transformation 08:58 – Power Up Program and Its Impact 19:23 – Guidance Center and Interactive Webinars 26:30 – Engaging in Live Webinars 34:57 – Copilot Studio Kit and Enterprise Impact Links & Resources Power Up – Free AI learning https://powerup.microsoft.com Agent Academy – Hands-on agent curriculum https://aka.ms/agent-academy Guidance Center – Enterprise implementation guidance https://learn.microsoft.com/microsoft-copilot-studio/guidance/ Architecture Center – Real-world agent architectures https://learn.microsoft.com/power-platform/architecture/products/copilot-studio AI Maturity & Transformation Playbook https://msft.it/6045sFYqh Copilot Studio Implementation Guide https://aka.ms/CopilotStudioImplementationGuide Copilot Studio Kit (Open source) https://github.com/microsoft/Power-CAT-Copilot-Studio-Kit CAT Team Blog – The Custom Engine https://microsoft.github.io/mcscatblog/ CAT Resources Portal https://microsoft.github.io/cat/ Power Platform YouTube Channel https://www.youtube.com/@mspowerplatform Power CAT AI Webinars Playlist https://www.youtube.com/playlist?list=PLnRSm4MRK-qdelQTsKXTXAIgO_PuGzSWw Copilot Studio Architecture Bootcamp https://microsoft.github.io/cat/programs/architecture-bootcamp.html Copilot Studio Resources Hub https://aka.ms/copilotstudio/resources
The conversation covers the introduction of the guest, John Seifert, and his role as the CEO of Dynamic Communities. It delves into the events organized by Dynamic Communities, including the Community Summit North America and the AI Agent and Co-Pilot Summit. The discussion also explores the concept of rapid prototyping with NLP, the activity spectrum, and the proactive skill development required in the AI space. Additionally, it addresses the importance of choosing the right AI tools for specific use cases. The conversation delves into the concept of using a diverse set of tools, akin to Batman's tool belt, to solve problems effectively. It emphasizes the importance of adopting and adapting AI tools to specific needs, as well as the value of technical understanding and creativity in leveraging these tools.