AI Orchestration with Semantic Kernel

Semantic Kernel is an open-source SDK that combines AI services such as OpenAI, Azure OpenAI, and Hugging Face, enabling developers to create AI apps by integrating them with their existing code. It supports Java, Python, and C# and offers connectors for adding memories and AI services to simulate the "brain" of the application. Additionally, AI plugins can be added as the "body" of the app. The SDK simplifies the integration of AI services, allowing developers to leverage recent AI advancements and build intelligent pipelines. This workshop focuses on how to use Semantic Kernel within your existing application and how to build and orchestrate AI applications.

Organizing your cloud-native Java backend

Are you grappling with the complexities of managing multiple microservices and shared libraries in your projects? Do you find yourself constantly in search of standards and rules for optimal repository management and release strategies? Are you looking to adopt a contract-first approach to enhance service interactions and streamline API definition releases and sharing? Furthermore, are you faced with the challenge of effectively managing configuration properties across different environments?

Let me share my experience implementing large-scale projects with Java/Spring backend with multiple DBs and Kafka as the heart of the application(s). Join me as I reveal proven strategies, essential principles, and real-world examples regarding Maven, the GitHub workflow/GitLab pipeline, service-to-service communication standards, the release and consumption of API definitions, and—most importantly—the administration and deployment of functional and technical configuration between cloud environments.

RAG-atouille

Retrieval Augmented Generation (RAG), similar to the artistry of Remy’s Ratatouille, combines the brilliance of Large Language Models (LLMs) with the precision of information retrieval. Just as Remy layers flavors in his dish, RAG-fusion seamlessly blends vectorized documents, images, audio, and video to craft nuanced responses in AI-powered applications. The RAG-multi index, like Gusteau’s secret spice blend, optimizes data organization, allowing LLMs to access a rich pantry of knowledge. And much like Anton Ego’s discerning palate, RAG-search ranking ensures that the most relevant insights rise to the top. Vector DB, our culinary laboratory, refines this recipe for a delectable user experience.

This presentation focuses on the latest architectural pattern called Retrieval Augmented Generation (RAG). I’ll start with a beginner-friendly introduction to why RAG is essential. Then, we’ll dive into practical implementation and design considerations, leveraging various data stores—both vectorized and non-vectorized. Finally, I’ll explore different RAG variations, including RAG-fusion, multi-index, and search ranking. Throughout, I’ll share real-world examples from my own experiences working with diverse customers and applications in this field, all of this with a pinch of "Ratatouille" (as the movie).

Beyond Bots: How to leverage power of AI

In this comprehensive exploration of AI's capabilities, I want to go beyond the traditional use of bots to unveil the full potential of AI in modern application development. A motivational session tailored for developers and architects, where I want to dive into diverse AI use cases and architecture diagrams across various domains & industries, including finance, banking, customer engagement, developer productivity, and retail. AI can be strategically implemented to revolutionize these sectors, offering a unique perspective on building robust, scalable solutions. Whether you're looking to enhance user experiences, streamline workflows, or drive innovation, this session will provide the knowledge and inspiration to leverage AI to its fullest potential.

I want to highlight innovative strategies for integrating AI to enhance user engagement, personalize experiences, and optimize operational efficiency. I'll share insights on the latest AI trends that are redefining industry benchmarks and how developers can harness these advancements to build cutting-edge applications. AI is not just an auxiliary tool but a central component in driving the future of technology, fostering a new era of intelligent, adaptive, and responsive applications that cater to the evolving needs of users and businesses alike.
This talk is close to my day-to-day job, where I run ideation workshop with Startups and Scaleups in Netherlands, to open up the minds on how to think AI beyond ChatGPT and bots.

LLMOps: A Developer’s Roadmap from Model to Production

In the rapidly evolving landscape of LLMs, architects and developers face the challenge of effectively selecting a model, deploying and managing applications based on language models (LLMs or SLMs) like GPT-4/Gemini/Phi3 etc. This talk focuses into the intricacies of operationalizing LLMs, focusing on prompt engineering, fine-tuning, and deployment strategies. Talk will explore the transformative potential of LLMs and address the critical aspects of bias mitigation, ethical considerations, and risk management. And will give insights into streamlining the LLM lifecycle, from model discovery to prompt tuning, ensuring domain-specific grounding, leveraging advanced optimization for fine-tuning and managing at production.

The session aims to equip attendees with a comprehensive understanding of the processes and best practices for deploying, managing LLMs/SLMs at scale, while prioritizing safety and responsibility.