Abstract
Rafael Pierre guides participants through the complete lifecycle of building production-ready agentic AI systems for enterprise banking. Drawing from a live pilot deployment using OpenAI Agents SDK and Phoenix Arize, this workshop demystifies multi-agent orchestration, observability strategies, and deployment patterns. Attendees will architect a conversational loan application system, implement agent coordination, and establish monitoring frameworks that satisfy enterprise requirements. This technical session bridges the gap between proof-of-concept demos and scalable production systems through real-world banking use cases.
Topics To Be Covered
Architect multi-agent systems effectively
Implement enterprise observability patterns
Debug agent interactions systematically
Design conversation flows strategically
Perfect For
Product Managers & Product Owners
AI/ML Engineers & Developers
IT Decision Makers & Enterprise Architects
Data & Platform Engineers
Meet Your Instructor
Rafael Pierre
Founder, Principal AI Engineer, weet.ai
Rafael Pierre is a Principal AI Engineer with a rare blend of experience across both AI hyperscalers and enterprise financial institutions. Having held key roles at Hugging Face and Databricks, he has deep expertise in large-scale AI infrastructure and open-source ecosystems. Rafael has also led the implementation of production-grade AI systems at major financial institutions such as ING Bank and ABN Amro, equipping him with practical insights into deploying conversational AI at scale in complex, regulated environments.
He specializes in advanced AI agent architectures, including Model Context Protocol (MCP), OpenAI Agents SDK, and session management systems designed for long-term memory in multi-agent environments. This powerful combination of hyperscaler innovation and real-world enterprise deployment positions Rafael as a unique voice in bridging cutting-edge AI capabilities with practical business outcomes.
ADDITIONAL INFORMATION
Time & Place
October 30, 2025
13:30 - 15:00
Hôtel Mövenpick Amsterdam City Centre
Matterhorn I
Limited to 45 participants.
Secure your seat – registration required.
Notes
Agenda for this session:
Getting Settled - 5 mins
Information Session - 40 mins
Break - 5 min
Individual / Group Exercise - 20 min
Q&A/Discussion - 20 min
Reflection - 5 min
Prerequisits:
Basic understanding of LLMs and conversational AI
Familiarity with API-based development and Python
OpenAI API Key
Laptop with Python 3.13 and Docker installed
VSCode
Git for cloning starter repository


.png)