video1.0<iframe src="https://www.loom.com/embed/4bf1c37ccd8f425286618eaecce5d2ba" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/4bf1c37ccd8f425286618eaecce5d2ba-4ea087877e3095bb.gif477.256In the video, I explain how Cody retrieves context from different sources, including local and remote repositories. I discuss the importance of providing specific technical terms and references for effective keyword-based context retrieval. I also touch on the use of BM25 for information retrieval and the decision to prioritize it over embeddings. Viewers are encouraged to engage in an iterative conversation with the model for better results.