AI and LLM strategies and application at Mercari Search

Mercari is a Japan-based second-hand e-commerce marketplace. We have been relying on Elastic Search for retrieval and DNN Learning to Rank for ranking for a long time. With the development of deep learning and LLM, In this talk, I would like to share we re-architecture our search system and convince internal stakeholders to take on new technology. In addition, how some of those new technologies work and not work in C2Ce-commerce setting.

1. End-to-end search architecture walkthrough and what new components are working for us
2. Success LLM application: offline dataset labeling, query expansions, and language translation using LLM and image embedding
3. Business consideration of the new technology such as ROI analysis”

Kaiyi Liu

Mercari

Kaiyi Liu is currently the Head of Engineering in Search and Discovery at Mercari in Tokyo, Japan. Mercari is a C2C marketplace where individuals can enjoy buying and selling items through the unique payment deposit system and use of AI technology. Kaiyi is leading initiatives that optimize user item match and facilitate transactions. For example, the team recently adopted GPT-4 and image search using SigLIP embeddings, Prior to Mercari, Kaiyi was a Senior Data Science Manager at TikTok, helping ensure content safety globally. Kaiyi received a master degree in Information Systems at Carnegie Mellon University.