Unleashing the Power of Vector Search and Semantic Caching in Java
summary
Spring Boot demo uses Redis OM for Java with the Deep Java Library (DJL) to vectorize images, store ~116,000 image embeddings (≈512-d floats) and run vector similarity queries with inner product and cosine similarity, indexed via FLAT or HNSW. The image face-match app returns top-N results (six–seven lines of query code), integrates with OpenAI/Hugging Face/LangChain/LlamaIndex, and implements semantic caching to cut LLM calls; the live demo works but the final stage went offline.
AI-written summary · about
chapters (13)
- 0:00Vector search and semantic caching
- 1:10Vector embeddings and use cases
- 5:31Vector databases in Redis
- 9:26Indexing algorithms and metrics
- 12:23Generative AI with RedisOM
- 13:29Demo setup Spring Boot RedisOM
- 17:47Redis data structures for embeddings
- 19:16Face detection demo
- 22:19Vector similarity in Java
- 24:45Dependencies and API overview
- 25:53AI use cases in Redis
- 26:35Redis vector search recap
- 27:38Q&A
related talks
55mVector Similarity Search in Spring with Redis Stack
01h 01mVector Similarity Search in Spring
01h 01mVector Similarity Search in Spring with Redis Stack
45mLearning Modern Java the Playful Way
52mLearning modern Java the playful way
45mLearning modern Java the playful way
59mLearning modern Java the playful way
50mA fun and absurd introduction to Vector Databases