This talk will show how to boost developer productivity with AI tools in Cursor — from automated code reading and writing to building custom AI teammates and end-to-end feature pipelines.

Half a year of trial and error with AI across different tasks and technologies: we’ll go through current tools, approaches, and recommendations. What works and what doesn’t. From the wow-effect in demos to wasted days on 'vibecoding'.

How can you thrive in an industry transformed by AI? This talk covers research and hands-on demos to help secure your career and leverage AI for productivity and growth.

Launching a Diia AI Against All Odds What does it take to build a conversational AI assistant for millions of citizens? The journey of creating Diia.AI, Ukraine's national digital assistant, started with a simple, promising Proof of Concept. But moving from a controlled demo to a live, production system that handles sensitive data and real-life services was a journey filled with unexpected turbulence. This talk is an honest, behind-the-scenes look at our architectural evolution. We'll dive deep into the real-world challenges we faced: from tackling unpredictable LLM hallucinations and integrating with complex legacy government registries, to designing for national-level security and scale under extreme pressure. This isn't a story of seamless success; it's a story of problems solved, from the absurdly simple to the monumentally complex. Join us to learn the practical lessons that aren't in the textbooks. We will cover how we architected for resilience using a RAG-based approach to fight misinformation, implemented robust Guardrails to ensure safety, and built a scalable, fault-tolerant ecosystem. This session is for any architect, developer, or product leader who wants to understand the battle scars and hard-won insights that come from launching a massive AI platform against all odds.

This talk presents an architectural approach to AI quality assurance in which a second AI system plays the role of a “judge,” replacing traditional testing methods. Using our real-world AI Calories Tracker an application that recognizes 5,000+ food photos every day, we will demonstrate how this unconventional strategy lets us manage risk effectively and boost trust in artificial intelligence.

We will analyze what actually works in the field of autonomous LLM agents, the existing limitations and their causes, research directions aimed at extending the duration of fully autonomous task execution, explore potential architectures of fully autonomous systems, economic autonomy of agents, and where we currently stand on the path toward achieving it.

Microsoft’s Vision. An overview of supply and demand, key trends in the development of artificial intelligence, as well as challenges in the corporate and public sectors.

Let's talk about the current state of AI investments, which products are attracting funding, whether it's a good time to invest now, and how to go about it.

We'll explore the key challenges of developing RAG-systems for the Ukrainian language. We'll discuss evaluation methods for such systems in specific domain areas, like consulting citizens on public services, and uncover practical insights and common "gotchas" that arise during implementation.