<?xml version="1.0" encoding="UTF-8"?><oembed><type>video</type><version>1.0</version><html>&lt;iframe src=&quot;https://www.loom.com/embed/90c2305905364629a5592c8bf0f1cfe3&quot; frameborder=&quot;0&quot; width=&quot;2064&quot; height=&quot;1548&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1548</height><width>2064</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1548</thumbnail_height><thumbnail_width>2064</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/90c2305905364629a5592c8bf0f1cfe3-2f60964cc56b4251.gif</thumbnail_url><duration>287.188</duration><title>Built-in AI Dashboards for Finance Insights</title><description>This Loom demonstrates Batuta, a built-in dashboard and query tool combining voice and string section capabilities. It shows a 1,000-row dataset with summaries like total value and key issues, including rule sets for basics of SAP, three-way match, duplication, and accountancy. Without using an LLM, it calculates total late payment penalty of 250,611.54 pence and can list duplicate invoices quickly. It then uses an LLM with the full file to produce insights, reporting 7 pence cost for the summarization and highlighting 152 blocked, 348 open items, 500 paid invoices, and top vendors by late payment exposure.</description></oembed>