Artificial Intelligence is no longer confined to digital spaces - it is rapidly expanding into the physical world, transforming how machines perceive, move, and interact with their environments. One may say that Robotics is the next frontier for AI to solve. You will get a perspective on developing in the field from a research engineer who’s been in the field near a decade and from an ML engineer who switched to the field just two years ago. We will talk about what Physical AI is, how it is described, advertised and what tasks it tries to solve. But also we will talk about how brittle and fragile things in the real world are and what unique challenges merging AI and robotics pose.

Most companies start their AI transformation with technology, not a business problem — and end up stuck in endless pilots instead of real results. In this talk, I’ll share practical principles shaped by building and delivering 20+ AI strategies and 50+ solutions across industries — from retail to pharma — in the US, Europe, and Ukraine. We’ll cover the most expensive mistakes, what actually drives ROI, and why some AI projects scale while others remain just slides. Who is this for: • CEOs, business owners, and leaders who are implementing or planning AI • Tech leads and ML engineers who want their AI solutions to deliver real business impact

What happens when a large language model becomes the entry point to government services that operate under real-world load and in the context of an information war? In such an architecture, any request may be not only incorrect but also intentionally manipulative — and standard AI safety solutions prove far less reliable than they appear in laboratory benchmarks. In this talk, I will share how we built a custom guardrail module for Diia.AI after encountering the limitations of off-the-shelf filters and the high cost of the LLM-as-a-Judge approach. Instead of validating every request with a large model, we designed a cascade security architecture: fast ML classifiers filter out most of the traffic, while the LLM is invoked only where deeper contextual analysis is truly required. This talk is not about perfect models, but about trade-offs, constraints, and practical decisions that must be made when an AI system operates not on a laptop, but within a national-scale service.

AI development hits everyone, so hit us. Everybody wants AI agents to replace regular UIs. In this talk, I will describe the evolution of our multitool AI agent, built with Node.js on top of Google Vertex AI. I’ll dive into our journey of choosing the right models and scaling development through CI/CD, TDD, and performance monitoring. Is it even possible to achieve stable results for AI projects that can hallucinate and return various responses? Interestingly, we eventually decided to remove MCP servers and Zod schema validation—technologies often considered the "standard" for these tasks. Want to know why we moved away from them? Join my session to get these insights and ask your questions live!

<p>MLOps (Machine Learning Operations) is a recent buzzword, that trends a lot. Let's figure out together how maintaining applications with machine learning components is significantly different from maintaining applications without them.</p> <p>We will look into MLOps best practices and typical problems and their implementations/solutions in real world production.</p>