November 25-27

AI agents are already causing real financial damage, and insurers are beginning to price that risk, with policies covering up to $50 million in agent-caused losses. As agent insurance becomes as common as cyber insurance, the key question is no longer “does it work?” but “is it insurable?”
Most teams unknowingly disqualify their systems from certification during the design phase. This talk explains why and shows how to avoid it.
Ioana Ghita presents a practical framework for designing AI agents that can be certified and insured. Drawing from real production deployments, she shows how five critical early architectural decisions determine whether an agent can meet emerging standards like AIUC-1.
Attendees will hear:
Five design decisions that make an agent certifiable
Four behavioral validations that generate auditable evidence
Three runtime metrics required to maintain certification

Each point is grounded in production failures where the architecture was the root cause, not the model.

“AI is transforming industries—and document processing is no exception. Modern AI systems take extraction to a new level of precision, drastically reduce training effort, and enable idea-to-production cycles within a single business day.

In this session, Eduard Shlepetskyy shows how organizations can move beyond templates and rule-based systems toward adaptive, learning-driven document intelligence. Through a live demonstration, attendees will see how structured data can be extracted from complex, unstructured documents in minutes—without lengthy setup or heavy IT projects.

The talk also addresses enterprise essentials such as accuracy, governance, and human validation, providing a practical framework for scaling AI-powered document processing sustainably.”

Orbit is transforming data consumption by replacing dashboards with agentic, context-aware AI. Rui Machado will explain how Orbit’s architecture, which integrates Orbie (a conversational agent), Heimdall (a security and access engine), and the Constellation API (an integration hub), allows users to obtain trusted, governed insights through simple queries. This session will demonstrate how Orbit streamlines traditional business intelligence (BI), automates analytical reasoning, and provides each employee with a personalised data assistant. Attendees will discover why agentic AI represents the next step in data consumption and how Orbit enables secure, scalable, and accessible data for all.

As AI agents become more capable, many engineering teams embed increasingly complex business logic directly into system prompts. While effective in prototypes, this approach creates invisible technical debt: nondeterministic failures, brittle model upgrades, and regression cycles that are difficult to reason about or validate.

In this technical case study, Stanko Kuveljić present the refactoring of a production scheduling agent from prompt-encoded logic to a deterministic, state-driven architecture. By separating model reasoning from application state transitions, the team introduced explicit invariant enforcement, structured observability, and behavioral testing gates across the continuous integration and continuous delivery (CI/CD) pipeline.

Key topics include:

Designing state graphs that enforce deterministic routing for business-critical operations

Implementing behavioral validation pipelines that test agent tool calls and state mutations instead of conversational phrasing

Reducing prompt surface area to improve traceability, reproducibility, and model upgrade safety

Introducing deployment gates that reduced manual regression validation effort in production

Attendees will leave with a reference architecture, a behavioral testing checklist, and a step-by-step approach for migrating prompt-driven agent systems toward bounded, production-grade reliability.

Liisa Thomas explores how behavioral science and organizational change principles are shaping the future of corporate compliance. In the presentation, she will share practical tools—like stakeholder mapping, influence planning, and cultural diagnostics—that privacy, AI and data professionals can use to integrate the power of organizational change into their compliance efforts. She will discuss how professionals can move past checklists and instead implement real, sustainable change in their organizations.

Generative AI is opening new possibilities for creating and transforming video in real time. In this talk, the speaker explores how recent models such as StreamDiffusion and LongLive push diffusion techniques into practical use for low-latency video generation and transformation. The speaker provides a deep technical walkthrough of how these systems can be adapted for streaming use cases, unpacking the full pipeline – from decoding, through the diffusion process, to encoding – and highlighting optimisation strategies, such as key–value (KV) caching, that make interactive generation possible. The talk also discusses the trade-offs between ultra-low-latency video transformation and generating longer, more coherent streams. To make it concrete, the speaker presents demos of StreamDiffusion (served with the open-source cloud service Daydream) and LongLive (explored with the open-source research tool Scope), showcasing practical examples of both video-to-video transformation and streaming text-to-video generation.

This session examines the emergence of autonomous AI agents that manage executive tasks such as scheduling, coordination, and routine decision-making. In this session, Aistis Raudys explains how these AI assistants can enhance organizational efficiency and redefine leadership workflows.