Could the real competitive advantage in the AI era come not from replacing human work, but from amplifying it? As artificial intelligence rapidly evolves, are professionals focusing enough on where their judgment and expertise matter?

In this episode of The Agency Blueprint podcast, I am joined by Thanos Diacakis to explore AI as a powerful productivity partner that enables agencies and knowledge workers to move faster while maintaining high-quality output. Thanos is a technology leader with over 25 years of experience helping startups and high-growth companies overcome software bottlenecks, scale engineering teams, and deliver high-quality products faster. He's held leadership roles at places like Uber, and today he's deeply involved in how teams are leveraging AI to do better, faster, more human-centered work.

Listen in to learn why AI still requires human judgment and expertise, especially when it comes to decision-making, prioritization, and contextual reasoning. You will also learn more about managing context in AI conversations, structuring prompts, experimenting with different AI tools, and building agent-based processes.

Key Questions:

[05:00] Is fear of being replaced preventing your team from embracing AI?
[13:41] How well do you understand how AI tools process context and information, and could that understanding help you use them more effectively?
[20:25] If you were managing a younger version of yourself, how would you structure the work with AI?
[22:46] Are you investing enough time experimenting with AI tools?

What You’ll Discover:

[01:28] Understanding what AI is actually good at today and where it still struggles in real-world workflows.
[03:10] How improvements in AI reliability are making multi-step automated workflows more effective than they were even months ago.
[05:41] Why AI still requires specialized human expertise rather than replacing professionals outright.
[09:07] How AI is forcing professionals to rethink their unique value and identify the skills that truly differentiate them.
[11:37] The possibility that AI could eventually replicate certain aspects of human reasoning by modeling patterns of thought.
[14:04] How AI conversations work through context and why understanding this is essential for effective prompting.
[18:38] How breaking tasks into smaller steps can significantly improve the reliability of AI-generated results.
[21:17] A practical framework for designing AI workflows by imagining how you would manage a junior employee performing the task.
[23:29] Thanos’ strategy for continuously testing new AI tools and reinvesting the time savings AI provides.
[25:00] Why beginners should dedicate a small percentage of their work time to learning and experimenting with AI.
[26:44] Why businesses that continuously improve their workflows with AI will quickly outpace competitors that do not.
[28:42] How fear of AI often comes from not understanding it and why experimentation reduces that fear.
[31:18] Why modern AI tools often require far less prompt engineering than people expect.
[32:21] MCP (Model Context Protocol) and how AI can use external tools to perform tasks like calculations or data retrieval.

Connect with Thanos: