AI BriefingMUSINSA develops DeciML and NeXA to move beyond prediction-centric personalization toward decision-making and discovery
·2026.11.03 09:20
Key point
MUSINSA has developed the DeciML and NeXA platforms to handle decision-making and discovery beyond prediction.
Details
- The MUSINSA Core-E Personalization team developed two ML platforms, DeciML and NeXA, to overcome the limitations of prediction
- DeciML addresses the problem of determining who to allocate the service's finite resources (such as push notifications and coupon budgets) to, while NeXA handles the problem of helping customers discover what they want
- It standardizes a three-stage process of understanding customers and products (Understanding), evaluating value (Valuation), and making selections considering constraints (Selection), designed so that the platform handles recurring technical issues
- It adopts an AI-native structure where humans and AI agents work under the same rules, accumulating the decision-making process and knowledge in the platform to use as a starting point for solving the next problem
- It aims to shorten the time it takes for DS and MLE to define and solve problems by standardizing the path from serving to deployment and operations
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
Musinsa Boosts Personalized Push CTR by 21.3% with Unified EmbeddingsMushinseok Developer: The Limit of Parallel AI Agent Work Is Judgment, Not CodingAAirbnb Introduces Proximity Features for Cold-Start Personalization Without User Identity
Figma's unreleased editing model, post-trained on Qwen 3.8 27B, beats top third-party model on internal evals