Building Scalable Cross-Modal Search with Ray

Learn how to build a state-of-the-art fashion image retrieval system in this lightning talk from Ray Summit 2024. Discover deepsense.ai's innovative approach to cross-modal search that combines Anyscale Ray and Pinecone technologies. Follow along as speakers Artur Zygadlo and Natalia Czerep break down the system architecture that enables seamless text-to-image and image-to-image queries using CLIP embeddings and vector similarity search. Gain practical insights into implementing autoscaling with Ray Serve and advanced parallelization techniques using Ray Data. Master the essential components needed to develop high-performance, user-friendly cross-modal search applications that can transform the e-commerce experience and extend beyond retail applications.