<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Blooper blog</title><description>Stories, research, notes, and news from the team building Blooper — the agentic AI filmmaking platform.</description><link>https://blooper.ai/</link><item><title>Grounding references in multi-image prompts</title><link>https://blooper.ai/blog/research/grounding-references/</link><guid isPermaLink="true">https://blooper.ai/blog/research/grounding-references/</guid><description>A one-line prompt rewrite that moves identity from 90% to 98% across every image provider — what we learned running 50 scenarios.</description><pubDate>Tue, 23 Jun 2026 00:00:00 GMT</pubDate><category>research</category><category>grounding</category><category>multi-image</category><category>nano-banana</category><category>openai</category><category>flux</category><author>Blooper Team</author></item><item><title>We launched a blog</title><link>https://blooper.ai/blog/news/blog-launch/</link><guid isPermaLink="true">https://blooper.ai/blog/news/blog-launch/</guid><description>Blooper now has a blog at /blog — stories, research, notes, and news. Four tracks, one feed.</description><pubDate>Tue, 23 Jun 2026 00:00:00 GMT</pubDate><category>news</category><category>launch</category><author>Blooper Team</author></item></channel></rss>