<?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/13d3cf12fd98414197247696a3401321&quot; frameborder=&quot;0&quot; width=&quot;1920&quot; height=&quot;1440&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1440</height><width>1920</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1440</thumbnail_height><thumbnail_width>1920</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/13d3cf12fd98414197247696a3401321-2621939673ab2c25.gif</thumbnail_url><duration>90.079773</duration><title>Singapore Bank Statement and Receipt Extraction Workflow</title><description>This Loom demonstrates a Singapore bank statement and receipt extraction workflow starting from a Google Drive folder. When new PDFs or images are added, the system routes them to the correct extraction prompt and, for bank statements, extracts all transactions, checks running balance, and validates statement integrity. It categorizes transactions using AI against a Singapore chart of accounts for items such as revenue, expenses, bank charges, transfers, and reversals, supporting DBS, OCBC, UOB, and HSBC in SGD and multi-currency formats. High-confidence results go to a zero-ready import sheet, while uncertain items are sent to a separate review queue with reasons shown for human approval.</description></oembed>