<?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/1e0f73c195544dbfb5e217fb0bfa948d&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/1e0f73c195544dbfb5e217fb0bfa948d-240afac67f1833dd-full.jpg</thumbnail_url><duration>678.446</duration><title>Leveraging Claude Code for Synthetic Data Generation 🚀</title><description>In this video, I discuss how we utilize Claude Code for our synthetic data projects, emphasising the importance of proper architecture and iterative development. I share my experience using ChatGPT 5 to generate a detailed project plan that includes data generation instructions and internal validation processes. I highlight the significance of creating a fine-grained execution plan and the necessity of testing after each step to ensure everything works as intended. I encourage viewers to adopt a structured approach to feature creation while maintaining rigorous software engineering principles. Please consider implementing these practices in your own projects for better results.</description></oembed>