Temporal developer advocate Shy Ruparel discusses survey results on the gap between daily agent use and production deployment, and between trust in agent output and reported reliability issues. He distinguishes the survey’s respondents and observations from universal outcomes.
Ruparel describes the difficulty of tracking state and debuggingDebugging is the process of finding why software behaved incorrectly and checking that a correction addresses the cause. long-running workflows, particularly where approvals and external systems are involved. He argues that observing a failure is only part of the problem; teams also need a way to recover progress.
A synthetic payment example demonstrates a workflow with an AI-assisted fraud review and human approvalHuman-in-the-loop command approval requires a person to review and authorize a consequential AI action before it executes.. Ruparel stops workers to illustrate how recorded workflow history can preserve completed steps and allow processing to resumeDurable workflow execution preserves progress so a multi-step process can resume after interruptions using recorded completion state.. He presents Temporal as the platform used for the demonstration.
Watch on YouTube
