https://www.infoq.com InfoQ Generative AI News feed https://www.infoq.com/news/2026/05/anthropic-claude-code-postmortem/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=Generative+AI-news <img src="https://res.infoq.com/news/2026/05/anthropic-claude-code-postmortem/en/headerimage/generatedHeaderImage-1778491231246.jpg"/><p>Anthropic published a postmortem tracing six weeks of Claude Code quality complaints to three overlapping product-layer changes: a reasoning effort downgrade, a caching bug that progressively erased the model's own thinking, and a system prompt verbosity limit that caused a 3% quality drop. The API and model weights were unaffected. All issues were resolved April 20.</p> <i>By Steef-Jan Wiggers</i> Large language models Generative AI AI Architecture Code Quality Anthropic AI, ML & Data Engineering Development news Thu, 14 May 2026 09:16:00 GMT https://www.infoq.com/news/2026/05/anthropic-claude-code-postmortem/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=Generative+AI-news Steef-Jan Wiggers 2026-05-14T09:16:00Z /news/2026/05/anthropic-claude-code-postmortem/en