MongoDB is introducing the embedded scoring pattern in Atlas Search, which allows users to incorporate term-level boosting logic directly within documents, enhancing search relevance and precision beyond the standard BM25 model. This technique addresses the limitations of query-time boosting by embedding scoring logic alongside content, enabling dynamic and context-aware control over search results. The implementation involves designing the index with embeddedDocuments and structuring documents to contain indexed_terms with specific boost values. Although powerful, this method requires careful planning to manage index size and performance. Concurrently, MongoDB has announced that, effective November 10, 2025, Chirantan “CJ” Desai will succeed Dev Ittycheria as CEO. Desai, with extensive experience from ServiceNow and Cloudflare, is expected to lead MongoDB into its next growth phase, emphasizing the company's strategic positioning in AI and data-intensive applications. Ittycheria, who will remain on the Board, expressed confidence in Desai's capabilities and highlighted the importance of leadership transitions for MongoDB's continued success.