Uploaded November 2026 | Updated November 2026, 2 days ago
Pedro Rodriguez, Director of Engineering, ML at Airbnb, explains why his team lets agents answer abstain when they aren't sure and sends those cases to a human, since in high-stakes cases a confident wrong answer is one of the worst possible outcomes. He asks viewers to imagine an agent that approves loans or pays insurance claims, where an answer that looks right isn't enough and you need to explain why the agent reached it.
Captured at Interrupt New York City, the agent conference by LangChain.
#AIAgents #HumanInTheLoop #Airbnb #LangChain #Interrupt
Pedro Rodriguez, Director of Engineering, ML at Airbnb, explains why his team lets agents answer abstain when they aren't sure and sends those cases to a human, since in high-stakes cases a confident wrong answer is one of the worst possible outcomes. He asks viewers to imagine an agent that approves loans or pays insurance claims, where an answer that looks right isn't enough and you need to explain why the agent reached it.
Captured at Interrupt New York City, the agent conference by LangChain.
#AIAgents #HumanInTheLoop #Airbnb #LangChain #Interrupt










