AI Master Podcast

Have you ever added a validation rule to an Angular form and then had to repeat the same rule in the UI? I asked Kobi Hari why that happens and how Angular Signal Forms can help.We started with a simple example. If a field is required, how does the component know to show an asterisk? Then we took it further. Could the same component show a useful hint or gain a new feature without having to rewrite it?In this episode, we talk about:• What frustrated Kobi about Reactive Forms• Why Validators.required and Validators.min(10) work differently• How field metadata can help reusable components decide what to show• How Kobi adds new capabilities to form fields and where AI fits inI also asked Kobi one of my favourite questions. After 30 years in software development, what still makes him excited to try something new?Kobi, thank you for such a fun conversation. I never thought a chat about validation could make me this curious about what a form field can do!And if you build forms in Angular, I’d love to hear from you. What’s the one bit of form code you’d happily never write again?Host: Dariusz KalbarczykGuest: Kobi HariPodcast: AI Master PodcastJoin us at NG Poland: https://ng-poland.plExplore AI Poland: https://ai-poland.plMore podcasts, people and ideas: https://kalbarczyk.dev#AIMasterPodcast #Angular #SignalForms

What happens when an AI agent becomes the Game Master in a role-playing game?

In episode 15 of AI Master Podcast, I talk with Arnaud Jean, Developer Advocate at AWS, about building AI agents through a project that is both useful and a lot of fun: an AI Game Master.

We explore how agents can roll dice, check game rules, create characters and work together to tell a story. Along the way, Arnaud explains the ideas behind tools, MCP, A2A and RAG, and why learning these concepts matters far beyond games.

You will hear about:
• What makes an AI agent different from a chat with an LLM.
• When one agent is enough, and when specialized agents can help.
• The orchestrator pattern and the trade-offs between control and autonomy.
• Building with Strands Agents and connecting tools through MCP.
• Using a knowledge base to help an agent check the rules.
• Starting locally and exploring Amazon Bedrock and AgentCore.
• Testing agent behavior with traces, OpenTelemetry and MCP Inspector.
• Why strong software fundamentals still matter as AI tools change.

We also talk about board games, family and the joy of learning by building something a little crazy.

Arnaud, thank you for such a fun and thoughtful conversation!

AI Poland: https://ai-poland.pl
More podcasts, people and ideas: https://kalbarczyk.dev

Host: Dariusz Kalbarczyk
Guest: Arnaud Jean
Language: English

A chatbot can give you an answer. An AI agent can take action. What changes when we give software the tools to do things for us?In this episode of AI Master Podcast, I sit down with Nir Kaufman to talk about building agents with TypeScript and keeping people in control.We discuss the agent loop, clear tool descriptions, choosing models for different tasks, structured output, memory and context. We also explore RAG, middleware, MCP, and the design choices behind single-agent and multi-agent systems.One of my favourite examples involves a pizza. When should an agent ask a person for approval? Ideally, before the pizza goes into the oven. It is a simple way to think about actions that are difficult to undo.We also talk about evaluation and tracing, the mindset behind Nir's hands-on workshop, and why music, hiking and real human connections still matter in a world full of AI.Nir brings more than 20 years of software development experience and a real love of teaching and community. It was a pleasure to have him back for this conversation.Enjoy the episode, and let me know: what would you trust an AI agent to do on its own?Nir's paid workshop, Agent Development in Practice with TypeScript: A Deep Dive:https://ai-poland.pl/workshop/agent-development-in-practice-with-typescriptNir's free workshop, From Zero to Production Agent in 60 Minutes:https://ai-poland.pl/free-workshops/zero-to-production-agentExplore AI Poland:https://ai-poland.pl/More podcasts and stories from me:https://kalbarczyk.dev/Host: Dariusz KalbarczykGuest: Nir KaufmanLanguage: English

Building an AI agent is relatively easy. Building one that is reliable, secure, measurable, and useful in a large company is much harder.

In this episode of the AI Master Podcast, I talk with Greg Radzio, who has spent almost two years designing and deploying AI agents for real business projects, including a large enterprise platform.

We examine the complete path from identifying a business problem to running an agent in production. Greg explains why improving the speed of individual employees does not automatically improve the performance of the whole organization.

We discuss:

✅ How an AI agent differs from a chatbot
✅ How to select a business process that is suitable for automation
✅ Why many AI projects remain proofs of concept
✅ The architecture of a production AI agent
✅ Models, tools, context, memory, and orchestration
✅ The role of an agent harness
✅ Testing systems that can return different results for the same input
✅ Evaluations, quality metrics, and business KPIs
✅ Preventing hallucinations and low-quality output
✅ Permissions, data access, and security boundaries
✅ Human approval for important operations
✅ Logging, tracing, monitoring, and incident handling
✅ The difference between a local coding assistant and a production agent
✅ How to introduce agents without automating a broken process
✅ What a company should achieve during its first 90 days

We also consider a fundamental question: can AI repair an inefficient business process, or will it only make that process fail faster?

At the end, Greg explains what developers, technical leaders, and business teams should understand before giving an AI agent access to company systems, code, and data.

Greg will continue this topic at AI Poland 2026.

Join us at our 2026 conferences:

🔴 NG Poland: https://ng-poland.pl
🟡 JS Poland: https://js-poland.pl
🟣 AI Poland: https://ai-poland.pl

A special thank you to Michał Łukawski and the entire SYZYGY team.

SYZYGY is the official partner of this episode. Thank you for supporting the AI Master Podcast and helping us create valuable technical conversations for the community! 🙏

#AIAgents #AgenticAI #EnterpriseAI #LLM #SoftwareArchitecture #AIEngineering #AIMasterPodcast #AIPoland

https://ai-poland.pl/

In this episode of the AI Master Podcast, Dariusz Kalbarczyk sits down with Daniel Sogl — Software Architect at Thinktecture AG, international speaker, MVP, Angular expert, and one of the developers actively exploring the future of AI-powered software development.

This conversation dives deep into one of the biggest transformations happening in tech right now: the shift from traditional coding to AI-assisted development.

Together, Dariusz and Daniel explore how modern AI coding tools are fundamentally changing the way developers write software, design architectures, collaborate in teams, and think about productivity.

The episode covers:

• the biggest mindset shifts developers must make in the AI era,

• when AI tools stopped being “interesting experiments” and became essential daily tools,

• how solutions like GitHub Copilot, Cursor, Windsurf, Bolt, and v0 differ from one another,

• and how developers can decide which AI tools are actually worth adopting.

Daniel shares practical insights from real-world development workflows, explaining how he integrates AI assistants into projects while still maintaining high standards of architecture, software quality, and engineering discipline.

One of the central themes of the episode is the balance between speed and responsibility.

The discussion explores:

• how to decide which tasks should be automated and which still require human judgment,

• whether AI-generated code risks creating dependency among junior developers,

• how teams can preserve code quality when AI adoption differs between developers,

• and why critical thinking may become even more important in the age of AI coding assistants.

The conversation also tackles the increasingly popular concept of “Vibe Coding” — what it actually means, whether AI tools encourage more creativity and experimentation, and how development workflows may evolve over the next 3–5 years.

Daniel also speaks openly about the limitations of today’s AI systems:

• hallucinations,

• misleading suggestions,

• architectural inconsistencies,

• and the importance of developers remaining deeply aware of what the code is actually doing.

Beyond the technical side, the episode includes a more personal discussion about growth, productivity, motivation, continuous learning, travel, and staying inspired in one of the fastest-changing industries in history.

If you want to understand where software development is heading, how AI coding assistants are reshaping engineering culture, and what developers should do to stay relevant in the coming years — this episode delivers practical insights, honest opinions, and a fascinating look into the future of programming.

Guest:

Daniel Sogl

Host:

Dariusz Kalbarczyk

https://ai-poland.pl

In this episode of the AI Master Podcast, Dariusz Kalbarczyk sits down with Gus Martins — Product Manager at Google DeepMind in London, former contributor to the TensorFlow ecosystem, and someone deeply passionate about building AI products that truly work in the real world.

This is a conversation about much more than artificial intelligence.

It’s about turning ambitious ideas into products used by millions of people. It’s about balancing research with practicality, innovation with reliability, and cutting-edge machine learning with real user needs.

Gus shares his incredible journey from Brazil to London and explains how he found his way into the world of machine learning, product development, and one of the most influential AI organizations on the planet.

Together, Dariusz and Gus explore:

• what it was like working on TensorFlow, TensorFlow Lite, and TensorFlow Hub during the early growth of modern ML ecosystems,

• how AI-powered products are actually designed and shipped inside Google DeepMind,

• what Product Managers really do in AI-focused teams,

• how engineers, researchers, and PMs collaborate on advanced AI systems,

• and why building AI is often much harder than the hype suggests.

The episode also dives deep into the practical side of AI product development:

• how to evaluate AI models for quality, fairness, and safety,

• the most common mistakes teams make when building ML products,

• how to move from a cool demo to a scalable and reliable product,

• and what separates useful AI from AI that simply looks impressive on stage.

One of the most fascinating parts of the discussion is Gus’ perspective on the role of community and open source in the growth of machine learning. Having spent years contributing to tools used by developers around the world, he explains why sharing knowledge, accessibility, and developer ecosystems remain critical even in the era of massive AI breakthroughs.

The conversation also takes a more personal turn as Gus talks about Brazilian culture, the rapidly evolving tech scene in Brazil, work culture differences, motivation, career growth, and what keeps him inspired while working at the forefront of AI innovation.

If you’re interested in machine learning, AI product development, Google DeepMind, TensorFlow, product management, or simply want to understand how modern AI products are actually created behind the scenes — this episode is packed with practical insights, honest reflections, and inspiring stories from someone building the future of AI every day.

Guest:

Gus Martins

Host:

Dariusz Kalbarczyk

https://ai-poland.pl

In this episode of the AI Master Podcast, Dariusz Kalbarczyk sits down with Andres Gesteira — software engineer, Angular expert, and AI enthusiast based in Madrid — to explore one of the hottest topics in modern software engineering: AI-assisted development.

Together, they dive deep into the rapidly evolving world of AI coding tools, intelligent agents, and the future of programming itself. Andres shares his hands-on experience working with modern AI-powered development environments, including Cursor and terminal-based workflows, while discussing how AI is changing the daily life of developers at companies like Fever.

The conversation covers everything from supervised agents and AI-driven productivity to the real-world strengths and weaknesses of today’s most popular coding assistants. Andres explains the differences between tools currently dominating the market, discusses privacy and security concerns surrounding AI-assisted coding, and shares how these systems can both accelerate development and introduce entirely new challenges.

Dariusz and Andres also tackle bigger questions facing the tech industry:

• Will AI replace developers — or simply augment them?

• How will AI reshape hiring, onboarding, and teamwork?

• Which skills will become more valuable in the AI era?

• And what should developers do today to stay ahead of the curve?

Beyond AI, the episode also offers a more personal look into Andres’ mindset, habits, motivation, favorite ways to relax after coding, and the lesser-known skills that truly matter in building a successful career in tech.

If you’re curious about the future of software development, AI-powered programming, modern developer workflows, and how experienced engineers are adapting to this new era — this episode is packed with practical insights, honest opinions, and inspiring ideas.

In this episode of the AI Master Podcast, we explore how artificial intelligence is reshaping leadership, strategy, and responsibility in modern organizations. Our guest is Dr. Maciej Madziński, a Strategy and Digital Transformation expert, MBA IT Programme Director, and lecturer at Kozminski University in Warsaw.

With over 15 years of experience at the intersection of business, technology, and leadership, Dr. Madziński shares a thoughtful and practical perspective on how AI is changing not only how companies operate, but also what it truly means to lead. The conversation is inspired by his upcoming talk at AI Poland 2025 titled “Leadership in the Age of AI: Between Promise and Responsibility.”

We discuss how leaders perceive AI today, whether excitement or fear dominates boardroom conversations, and why treating AI as just another tool is one of the biggest misunderstandings. Dr. Madziński explains how AI is redefining strategy, where leaders struggle most during adoption, and what responsible AI implementation looks like in real organizational settings.

A strong focus of the episode is placed on human-centered leadership. We talk about empathy, trust, courage, and ethical responsibility in a world where algorithms increasingly support decision-making. The episode also touches on inequality, reskilling, and how leaders can motivate people to grow rather than fear being replaced by technology.

This episode is highly relevant for executives, managers, founders, technologists, MBA students, and anyone interested in leading people and organizations responsibly in the age of AI.

Learn more about our conferences and community:

AI Poland - https://ai-poland.pl

NG Poland - https://ng-poland.pl

JS Poland - https://js-poland.pl

🎙️ AI Master Podcast | HR in the Age of AI

In this episode of the AI Master Podcast, we dive into a topic that is becoming critical for modern organizations: HR in the age of artificial intelligence.

Our guest, Joanna Rozmysłowska, brings extensive experience in organizational culture, IT recruitment, and employer branding. She has worked closely with developers, leaders, and international teams, supporting growth, building confidence, and helping organizations navigate change in a fast-evolving tech landscape.

Based in Warsaw, Joanna shares a practical and human perspective on how AI is reshaping HR, from recruitment processes and leadership models to culture building in AI-driven workplaces.

🔍 In this episode, you’ll learn:

• what the core role of HR looks like today and how AI is transforming it

• the biggest challenges companies face when introducing AI into HR processes

• skills HR professionals need to stay relevant in the coming years

• how AI impacts recruitment, from first contact to final offer

• the balance between hard skills, soft skills, and a growth mindset

• leadership behaviors that truly matter in the age of AI

• why empathy and humanity remain essential in HR

• what’s actually working in employer branding as we move toward 2026

• Joanna’s personal perspective on her work, challenges, and what excites her most about the future of HR

🌍 Join our developer community and events to expand your learning, connect with peers, and meet world-class speakers, whether your focus is Angular, JavaScript, or AI.

🔗 NG Poland, the largest Angular conference in Europe

https://ng-poland.pl

🔗 JS Poland, the biggest JavaScript conference in CEE

https://js-poland.pl

🔗 AI Poland, a conference focused on practical AI applications

https://ai-poland.pl

#AI #HumanResources #AIinHR #HRTech #EmployerBranding #Leadership #TechPodcast #NGPoland #JSPoland #AIPoland

AI Master Podcast – Ondřej Chrastina: Building the Future of Content Editing with AI

In this episode of the AI Master Podcast, I’m joined by Ondřej Chrastina, Developer Advocate at CKEditor and an experienced Developer Relations leader with over 12 years of hands-on and leadership experience in building outstanding developer experience for complex platforms.

Ondřej shares his professional journey — from QA engineering and software development, through building internal tooling and SDKs, to establishing and leading Developer Relations programs at companies such as Kentico, Kontent.ai, Ataccama, and now CKEditor. We talk about how deep empathy for developer pain points shapes better products, tooling, and documentation.

A large part of our conversation focuses on CKEditor 5 — its modern, extensible architecture and the challenges of using rich-text editors in today’s web applications. We explore how generative AI and large language models can meaningfully enhance content editing, turning editors into intelligent assistants rather than passive tools.

Ondřej explains practical AI use cases such as content suggestions, summarization, auto-formatting, review workflows, and AI-assisted collaboration. We also go deep into the technical and product challenges of embedding AI into editors: latency, context management, hallucinations, prompt engineering, safety, and maintaining full user control — especially in enterprise environments where privacy, data governance, and GDPR compliance are critical.

We also discuss the future of rich-text editors in an AI-driven world, the role of open source and community contributions, multilingual challenges, and which AI advancements (embeddings, RAG, multimodal models) are most exciting for the next generation of editor experiences.

To wrap up, we switch to a more personal perspective — continuous learning, staying motivated as a software engineer, productivity, travel, and a few lesser-known facts about Ondřej.

⸻

Key topics covered in this episode:

• Architecting AI-powered features in rich-text editors while preserving user control

• Integrating LLMs into editors: latency, context size, and hallucination challenges

• Prompt engineering patterns for editor-based AI workflows

• Privacy, GDPR, and data governance in enterprise-grade AI integrations

• The future of rich-text editors as intelligent, collaborative assistants

Links mentioned in the episode:

CKEditor AI

Meet your app’s new AI co-writer. Easily integrated, full-featured AI writing assistant for your application.

Discover CKEditor AI features that help you write, edit, and improve content faster — including AI Chat with chat history, AI Quick Actions, and AI Review.

👉 https://ckeditor.com/ckeditor-5/capabilities/ai-features/?utm_campaign=ckeditor_podcast_ai_master_podcast&utm_source=podcast_description&utm_medium=referral

CKEditor Docs

Master CKEditor 5 — customize it, build powerful plugins, and create your own editors. Ask AI to get started faster!

👉 https://ckeditor.com/docs/ckeditor5?utm_campaign=ckeditor_podcast_ai_master_podcast&utm_source=podcast_description&utm_medium=referral

CKEditor Builder

Pick your preset (AI-recommended), choose your features, generate your editor, and start hacking away!

👉 https://ckeditor.com/ckeditor-5/builder/?utm_campaign=ckeditor_podcast_ai_master_podcast&utm_source=podcast_description&utm_medium=referral

Guest: Ondřej Chrastina

Developer Advocate, CKEditor

🌐 Website: https://ondrej.chrastina.dev/

💼 LinkedIn: https://www.linkedin.com/in/ondrejchrastina/

💻 GitHub: https://github.com/Simply007

This episode is a must-listen for developers, DevRel professionals, and product teams interested in AI-powered content creation, developer tooling, and the future of web editors.