https://nyghtowl.com/tags/reliability/ Recent content in Reliability on Nyghtowl Hugo en-us <a href="https://creativecommons.org/licenses/by-nc-sa/4.0/" target="_blank" rel="noopener">CC BY-NC-SA 4.0</a> Thu, 02 Apr 2026 12:59:04 +0000 https://nyghtowl.com/posts/2026/04/ai-reliability-is-a-decade-old-problem/ Thu, 02 Apr 2026 12:59:04 +0000 https://nyghtowl.com/posts/2026/04/ai-reliability-is-a-decade-old-problem/ <p>I recently contributed this <a href="https://temporal.io/blog/ai-reliability-is-a-decade-old-problem">piece to Temporal’s blog</a> and wanted to share it, too.</p> <hr> <p>The AI agents being deployed today can reason through complex tasks, chain together dozens of tool calls, and operate autonomously for hours. What most of them <em>can’t</em> do is survive something going wrong halfway through.</p> <p>Even if an agent were 85% reliable at each step, a 10-step workflow would succeed end-to-end only about 20% of the time. Scale that to the longer workflows that production agents actually run, and even strong step-level performance produces cascading failure. Not because the model got something wrong, but because the system had no way to checkpoint progress, recover from a partial failure, or resume where it left off.</p>