We Trust AI Agents That Break Every Day - Shy Ruparel, Temporal

Temporal's State of Development survey found that 80.8% of engineers use agents daily, but only 49.1% run them in production. Developer advocate Shy Ruparel walks through the four gaps behind that number (usage vs. production, trust vs. reality, prototype vs. enterprise, seeing vs. recovering), then demos a $2,500 payout workflow with a local model that survives having its workers killed mid-flight, without losing state, repeating tool calls or re-spending tokens. Chapters: 0:00 Intro 0:34 Finance runs on long-running workflows 2:05 The State of Development survey 3:16 Gap 1: Usage vs. production 4:33 Gap 2: Trust vs. reality 7:18 Gap 3: Prototype vs. enterprise 8:55 Gap 4: Seeing vs. recovering 10:53 Demo: a payout that survives a crash 13:06 Durable execution explained 18:12 What the demo solved 20:10 Who runs this in production 20:50 Wrap-up Shy Ruparel https://temporal.io Join us in SF for the second annual AIE CODE summit: https://aie.codes our last flagship summit of 2026. Applications live now!