AI Agents That Follow the Rules: Sandi Besen

Description

In this episode of the ODSC Ai X Podcast, host Sheamus McGovern speaks with Sandi Besen, AI Engineer and Ecosystem Lead at IBM Research, about designing rule-based, production-ready AI agents using IBM’s open-source BeeAI. Sandi shares her journey from the performing arts to AI engineering, unpacks the challenges of agent reliability and trust, and explores how frameworks like BeeAI are solving real-world issues through enforceable rules, observability, and open standards. This episode dives deep into multi-agent systems, communication protocols, and what it really takes to deploy AI agents safely and effectively in enterprise environments. Key Topics Covered: ▪️ Sandi’s background and current role at IBM Research ▪️ What inspired the creation of the BeeAI framework and platform ▪️ The concept of "agent chaos" and how BeeAI brings structure and reliability ▪️ Comparison of popular agent frameworks: LangChain, LangGraph, CrewAI ▪️ Introduction to emerging agent communication protocols (MCP and A2A) ▪️ Enforcing hardcoded rules (conditional requirements) in agents ▪️ Building agent behavior guardrails and preventing rogue actions ▪️ Importance of observability, memory, and tracing in agent frameworks ▪️ Using OpenTelemetry for tracing agent decision-making ▪️ Challenges in agent safety, validation, and trust for enterprise use ▪️ How BeeAI makes deploying an agent with a UI fast and easy (in under an hour) ▪️ When to build from scratch vs. using a framework like BeeAI  ▪️ Common mistakes startups make when adopting cutting-edge AI tech ▪️ Trends in context engineering and long-term memory ▪️ The future of consulting in the age of AI