Clayton Hotel Charlemont
**Reliable Agentic Systems on Snowfalke: A Cortex Agent Deep Dive, Anastasiia Stefanska, Snowflake Data Superhero** Building a Cortex Agent is easy. Building one that is reliable, observable, cost-effective and maintainable in production is a different challenge. In this session, we go beyond the standard Cortex Agent configuration and look at the engineering patterns that become important as agentic applications grow in complexity. We will explore how to evaluate agent behaviour rather than just final answers, how to use traces and AI observability to understand what an agent is actually doing, and how to monitor and control the cost of agent execution. We will then move beyond the individual agent and look at agentic harnesses and orchestration patterns - including when to rely on the Cortex Agent's built-in reasoning loop, when to introduce deterministic control around it, and how multiple specialised agents and tools can be combined into larger workflows. [Anastasiia Stefanska](https://www.linkedin.com/in/anastasiastefanska/) is a Berlin-based data leader, Snowflake Data Superhero, and co-founder of the Women on Snowflake user group. She works as a lead of a data team and specialises in managing data products using Snowflake and Tableau. She is an active speaker and contributor in the data community, recently sharing insights on Cortex Agents and presenting at Snowflake Summit events regarding AI-ready data, AI agents, and careers in the AI age. **Building Reliable AI Agents: Best Practices for Implementation & Governance, Dmytro Yaroshenko** [Dmytro Yaroshenko](https://www.linkedin.com/in/yarodmi/) is former Field CTO at Snowflake with extensive experience in agentic systems, AI governance, and enterprise data platforms. Dmytro focuses on helping teams design and deploy scalable, trustworthy AI agents in production environments. **Location** Clayton Hotel Charlemont **Agenda** * 6:00 pm – Food & Networking * 6:30 pm – Reliable Agentic Systems on Snowfa
Tuesday, October 6 · 1:00 PM