Rediscovering Apollo 11: Using Kotlin, Spring AI + Redis OM Spring to explore the mission to the moon!

What happens when you combine the Apollo program’s historical data with modern AI tools? You get a way to interact with one of humanity’s greatest adventures like never before!

In this session, I’ll show you how I used Redis OM Spring and Spring AI to explore Apollo mission data—aligning transcripts, telemetry, and images to uncover hidden connections and insights. We’ll dive into how Semantic Search powered by vector embeddings makes sense of unstructured text, how Redis as a vector database enables lightning-fast retrieval, and why these tools unlock new ways to explore complex datasets.

Don’t know what embeddings or vector databases are? No worries—I’ll break it all down and show you how it works.

Come for the Moon missions, stay for the AI magic, and leave ready to build your own intelligent search experiences!

1. Raphael De Lio REDISCOVERING APOLLO 11 Using Kotlin, Redis, and Vector Search
2. CONTEXT
3. OUR MISSION
4. OBJECTIVE "That's one small step for man, one giant leap for Mankind." "That's one small step for a human, one big jump for humanity.” AND FIND SEARCH FOR
5. VECTOR SIMILARITY SEARCH
6. Which one is the most similar to Earth?
7. 0 -500 500 298 152 50 45 12
8. -500 Jupiter (-110, 500) 0 -500 500 500 Temp Mass Displaying only Mercury, Venus, Jupiter, Saturn and Uranus for simpli fi cation. Mass normalized from -500 to 500 Mercury (167, -500) Venus (465, -497) Saturn (-178, -200) Uranus (-195, -454)
9. -500 Jupiter (-110, 500) 0 -500 500 500 Temp Mass Displaying only Mercury, Venus, Jupiter, Saturn and Uranus for simpli fi cation. Comparing how similar Saturn is to other planets in terms of mass and temperature. The shortest, the closest. Mercury (167, -500) Venus (465, -497) Saturn (-178, -200) Uranus (-195, -454) d(S, U) = 254 d(S, J) = 703 d(S, M) = 457 d(S, V) = 708
10. -500 Jupiter (-110, 500) 0 -500 500 500 Temp Mass Displaying only Mercury, Venus, Jupiter, Saturn and Uranus for simpli fi cation. Comparing how similar Saturn is to other planets in terms of mass and temperature. The lowest, the closest. Mercury (167, 500) Venus (465, -497) Saturn (-178, -200) Uranus (-195, -454) ∠(S, U) = 18°
11. -500 Jupiter (-110, 500) 0 -500 500 500 Temp Mass Displaying only Mercury, Venus, Jupiter, Saturn and Uranus for simpli fi cation. Comparing how similar Saturn is to other planets in terms of mass and temperature. The lowest, the closest. Mercury (167, -500) Venus (465, -497) Saturn (-178, -200) Uranus (-195, -454) ∠(S, M) = 60°
12. -500 Jupiter (-110, 500) 0 -500 500 500 Temp Mass Displaying only Mercury, Venus, Jupiter, Saturn and Uranus for simpli fi cation. Comparing how similar Saturn is to other planets in terms of mass and temperature. The lowest, the closest. Mercury (167, -500) Venus (465, -497) Saturn (-178, -200) Uranus (-195, -454) ∠(S, M) = 84°
13. -500 Jupiter (-110, 500) 0 -500 500 500 Temp Mass Displaying only Mercury, Venus, Jupiter, Saturn and Uranus for simpli fi cation. Comparing how similar Saturn is to other planets in terms of mass and temperature. The lowest, the closest. Mercury (167, -500) Venus (465, -497) Saturn (-178, -200) Uranus (-195, -454) ∠(S, M) = 84°
14. - A vector is a numerical representation of data - Vectors that are closer together in a vector space model are more similar to each other - Their proximity is calculated using euclidean distance or cosine similarity
15. WHERE DOES AI FIT?
16. [15, -497] [-178, -200]
18. [0.54, 0.86, 0.23, 0.75, 0.92, 0.64, 0.47, 0.33, 0.89, 0.99, 0.67, 0.49, 0.94, 0.36, 0.71, 0.82, 0.29, 0.57, 0.630.98, 0.64, 0.47, 0.33, 0.48, 0.95, 0.67, 0.81, 0.23, 0.21, 0.67, 0.49, 0.56, 0.57, 0.63, 0.89, 0.21, 0.67, 0.49, 0.94, 0.36, 0.71, 0.82, 0.29] Embedding Model
19. How does it work? VECTOR SIMILARITY SEARCH "That's one small step for man, one giant leap for Mankind." "That's one small step for a human, one big jump for humanity.” [0.56, 0.76, 0.80, 0.54, 0.99, -0.87 … ] [0.58, 0.75, 0.75, 0.63, 0.9, -0.7 … ]
20. Where to store our vectors? VECTOR SIMILARITY SEARCH 1 Billion Vectors - 90% precision - 200ms median latency 1 Billion Vectors - 95% precision - 1.3s median latency
21. - Embedding models are algorithms that transform unstructured data into meaningful numerical data - Vector databases are used for e ff i ciently storing and retrieving these vectors based on their proximity in the vector space model
22. BUILDING OUR ROCKET
23. How to interact with Redis? VECTOR SIMILARITY SEARCH Saturn V Stage 1: JVM Stage 2: Kotlin Stage 3: Spring Framework Command Module & Service: Redis OM Spring & RedisVL Lunar Module: Spring AI Escape System: Caching
24. The Command Module VECTOR SIMILARITY SEARCH How to interact with Redis? JSON System Vector System Query Engines Probabilistic Data Structures Support Perform ance Boosters DS Enhancers Redis OM Spring
25. Mission Control VECTOR SIMILARITY SEARCH How to interact with Redis?
26. • Collect our data • Load our data into Redis • Query our data ROCKET’S OBJECTIVE
27. COLLECTING THE DATA
28. LOADING THE DATA @RedisHash data class Utterance( @Id var timestamp: String, @Indexed var text: String, @Indexed @NonNull var speaker: String, @Indexed @NonNull var speakerId: String = "" )
29. @RedisHash data class Utterance( @Id var timestamp: String, @Indexed var timestampInt: Int = 0, @Indexed @NonNull var speaker: String, @Indexed @NonNull var speakerId: String = "" ) LOADING THE DATA Annotation for de fi ning a HASH object
30. @RedisHash data class Utterance( @Id var timestamp: String, @Indexed var text: String, @Indexed @NonNull var speaker: String, @Indexed @NonNull var speakerId: String = "" ) LOADING THE DATA Annotation for automatic creating a ULID
31. @RedisHash data class Utterance( @Id var timestamp: String, @Indexed var text: String, @Indexed @NonNull var speaker: String, @Indexed @NonNull var speakerId: String = "" ) LOADING THE DATA Annotation for creating indexes within Redis (Query Engine) for e ff i cient fi ltering
32. @RedisHash data class Utterance( @Id var timestamp: String, @Indexed var text: String, @Indexed @NonNull var speaker: String, @Indexed @NonNull var speakerId: String = "" ) LOADING THE DATA Annotation for creating indexes within Redis (Query Engine) for e ff i cient fi ltering
33. LOADING THE DATA Creating the repository @RedisHash data class Utterance( @Id var timestamp: String, @Indexed var text: String, @Indexed @NonNull var speaker: String, @Indexed @NonNull var speakerId: String = "" ) interface UtteranceRepository : RedisEnhancedRepository<Utterance, String>
34. LOADING THE DATA JSON instead of HASH interface UtteranceRepository : RedisDocumentRepository<Utterance, String> @Document data class Utterance( @Id var timestamp: String, @Indexed var text: String, @Indexed @NonNull var speaker: String, @Indexed @NonNull var speakerId: String = "" )
35. VECTORIZING THE DATA @NonNull @Vectorize( destination = "embeddedText", embeddingType = EmbeddingType.SENTENCE ) var text: String, @VectorIndexed( algorithm = VectorField.VectorAlgorithm.HNSW, dimension = 384, distanceMetric = DistanceMetric.COSINE ) var embeddedText: ByteArray? = null,
36. @NonNull @Vectorize( destination = "embeddedText", embeddingType = EmbeddingType.SENTENCE ) var text: String, @VectorIndexed( algorithm = VectorField.VectorAlgorithm.HNSW, dimension = 384, distanceMetric = DistanceMetric.COSINE ) var embeddedText: ByteArray? = null, VECTORIZING THE DATA Data to be vectorized
37. @NonNull @Vectorize( destination = "embeddedText", embeddingType = EmbeddingType.SENTENCE ) var text: String, @VectorIndexed( algorithm = VectorField.VectorAlgorithm.HNSW, dimension = 384, distanceMetric = DistanceMetric.COSINE ) var embeddedText: ByteArray? = null, VECTORIZING THE DATA Field to store the vector Index Con fi guration
38. @NonNull @Vectorize( destination = "embeddedText", embeddingType = EmbeddingType.SENTENCE ) var text: String, @VectorIndexed( algorithm = VectorField.VectorAlgorithm.HNSW, dimension = 384, distanceMetric = DistanceMetric.COSINE ) var embeddedText: ByteArray? = null, VECTORIZING THE DATA Flat or HNSW
39. @NonNull @Vectorize( destination = "embeddedText", embeddingType = EmbeddingType.SENTENCE ) var text: String, @VectorIndexed( algorithm = VectorField.VectorAlgorithm.HNSW, dimension = 384, distanceMetric = DistanceMetric.COSINE ) var embeddedText: ByteArray? = null, VECTORIZING THE DATA Number of dimensions created by the embedding model
40. @NonNull @Vectorize( destination = "embeddedText", embeddingType = EmbeddingType.SENTENCE ) var text: String, @VectorIndexed( algorithm = VectorField.VectorAlgorithm.HNSW, dimension = 384, distanceMetric = DistanceMetric.COSINE ) var embeddedText: ByteArray? = null, VECTORIZING THE DATA Cosine or Euclidean
41. VECTORIZING THE DATA Di ff erent embedding model Number of dimensions must match the number of dimensions of the model
42. @Vectorize( destination = "embeddedQuestion", embeddingType = EmbeddingType.SENTENCE, provider = EmbeddingProvider.OPENAI, openAiEmbeddingModel = OpenAiApi.EmbeddingModel.TEXT_EMBEDDING_3_LARGE ) var question: String, @VectorIndexed( algorithm = VectorAlgorithm.HNSW, dimension = 3072, distanceMetric = DistanceMetric.COSINE, ) var embeddedQuestion: ByteArray? = null VECTORIZING THE DATA Di ff erent embedding model Number of dimensions must match the number of dimensions of the model
43. UNDER THE HOOD Spinning up Transformers Embedding Model
44. UNDER THE HOOD Spinning up OpenAI Embedding Model
45. QUERY THE DATA This is going to generate the command to search and sort on Redis. This is done e ff i ciently by the Redis Query Engine private val entityStream: EntityStream private val embedder: Embedder val embedding = embedder.getTextEmbeddingsAsBytes(listOf(text), `Utterance$`.TEXT).first() val stream: SearchStream<Utterance> = entityStream.of(Utterance : : class.java) val results = stream .filter(`Utterance$`.EMBEDDED_TEXT.knn(3, embedding)) .sorted(`Utterance$`._EMBEDDED_TEXT_SCORE) .map(Fields.of(`Utterance$`._THIS, `Utterance$`._EMBEDDED_TEXT_SCORE)) .collect(Collectors.toList())
46. private val entityStream: EntityStream private val embedder: Embedder val embedding = embedder.getTextEmbeddingsAsBytes(listOf(text), `Utterance$`.TEXT).first() val stream: SearchStream<Utterance> = entityStream.of(Utterance : : class.java) val results = stream .filter(`Utterance$`.EMBEDDED_TEXT.knn(3, embedding)) .sorted(`Utterance$`._EMBEDDED_TEXT_SCORE) .map(Fields.of(`Utterance$`._THIS, `Utterance$`._EMBEDDED_TEXT_SCORE)) .collect(Collectors.toList()) QUERY THE DATA This is going to generate the command to search and sort on Redis. This is done e ff i ciently by the Redis Query Engine Provided by Redis OM Spring
47. private val entityStream: EntityStream private val embedder: Embedder val embedding = embedder.getTextEmbeddingsAsBytes(listOf(text), `Utterance$`.TEXT).first() val stream: SearchStream<Utterance> = entityStream.of(Utterance : : class.java) val results = stream .filter(`Utterance$`.EMBEDDED_TEXT.knn(3, embedding)) .sorted(`Utterance$`._EMBEDDED_TEXT_SCORE) .map(Fields.of(`Utterance$`._THIS, `Utterance$`._EMBEDDED_TEXT_SCORE)) .collect(Collectors.toList()) QUERY THE DATA This is going to generate the command to search and sort on Redis. This is done e ff i ciently by the Redis Query Engine Vectorizing user query
48. private val entityStream: EntityStream private val embedder: Embedder val embedding = embedder.getTextEmbeddingsAsBytes(listOf(text), `Utterance$`.TEXT).first() val stream: SearchStream<Utterance> = entityStream.of(Utterance : : class.java) val results = stream .filter(`Utterance$`.EMBEDDED_TEXT.knn(3, embedding)) .sorted(`Utterance$`._EMBEDDED_TEXT_SCORE) .map(Fields.of(`Utterance$`._THIS, `Utterance$`._EMBEDDED_TEXT_SCORE)) .collect(Collectors.toList()) QUERY THE DATA This is going to generate the command to search and sort on Redis. This is done e ff i ciently by the Redis Query Engine
49. private val entityStream: EntityStream private val embedder: Embedder val embedding = embedder.getTextEmbeddingsAsBytes(listOf(text), `Utterance$`.TEXT).first() val stream: SearchStream<Utterance> = entityStream.of(Utterance : : class.java) val results = stream .filter(`Utterance$`.EMBEDDED_TEXT.knn(3, embedding)) .sorted(`Utterance$`._EMBEDDED_TEXT_SCORE) .map(Fields.of(`Utterance$`._THIS, `Utterance$`._EMBEDDED_TEXT_SCORE)) .collect(Collectors.toList()) QUERY THE DATA This is going to generate the command to search and sort on Redis. This is done e ff i ciently by the Redis Query Engine
50. private val entityStream: EntityStream private val embedder: Embedder val embedding = embedder.getTextEmbeddingsAsBytes(listOf(text), `Utterance$`.TEXT).first() val stream: SearchStream<Utterance> = entityStream.of(Utterance : : class.java) val results = stream .filter(`Utterance$`.EMBEDDED_TEXT.knn(3, embedding)) .sorted(`Utterance$`._EMBEDDED_TEXT_SCORE) .map(Fields.of(`Utterance$`._THIS, `Utterance$`._EMBEDDED_TEXT_SCORE)) .collect(Collectors.toList()) QUERY THE DATA This is going to generate the command to search and sort on Redis. This is done e ff i ciently by the Redis Query Engine
51. LIFT OFF
52. HOUSTON, WE HAVE AN (ACCURACY) PROBLEM
53. CHUNKING?
54. • Fixed-size chunking • Markdown-based chunking • Semantic chunking CHUNKING?
55. • Short text: “Apollo” → Doesn’t tell if it’s about a Greek god, a space mission, or a music album. CHUNKING?
56. GROUPING
57. TABLE OF CONTENTS
58. TABLE OF CONTENTS
59. OTHER TECHNIQUES
60. SUMMARIZATION Related U tt erances LLM Embedding Model Summarize Vectorize Store
61. SUMMARIZATION @Bean fun summarizationChatClient( chatModel: ChatModel, ): ChatClient { return ChatClient.builder(chatModel) .defaultSystem(DEFAULT_PROMPT) .build() } companion object { private val DEFAULT_PROMPT = """ You are a helpful assistant who summarizes utterances of the Apollo 11 mission. Make these summaries very dense with all curiosities included. Limit the summary to 512 words. """.trimIndent() }
62. SUMMARIZATION val processed = tocList.map { toc -> async(Dispatchers.IO) { runCatching { logger.info("Generating summary for TOC entry: {}", toc.startDate) val response = summarizationChatClient .prompt() .user(toc.concatenatedUtterances.orEmpty()) .call() .chatResponse() toc.summary = response ?. result ?. output ?. text logger.info("Successfully generated summary for TOC entry: {}", toc.startDate) toc }.onFailure { e -> logger.error("Error generating summary for TOC entry: {}", toc.startDate, e) }.getOrNull() } }.mapNotNull { it.await() }
63. SUMMARIZATION val processed = tocList.map { toc -> async(Dispatchers.IO) { runCatching { logger.info("Generating summary for TOC entry: {}", toc.startDate) val response = summarizationChatClient .prompt() .user(toc.concatenatedUtterances.orEmpty()) .call() .chatResponse() toc.summary = response ?. result ?. output ?. text logger.info("Successfully generated summary for TOC entry: {}", toc.startDate) toc }.onFailure { e -> logger.error("Error generating summary for TOC entry: {}", toc.startDate, e) }.getOrNull() } }.mapNotNull { it.await() }
64. SUMMARIZATION val processed = tocList.map { toc -> async(Dispatchers.IO) { runCatching { logger.info("Generating summary for TOC entry: {}", toc.startDate) val response = summarizationChatClient .prompt() .user(toc.concatenatedUtterances.orEmpty()) .call() .chatResponse() toc.summary = response ?. result ?. output ?. text logger.info("Successfully generated summary for TOC entry: {}", toc.startDate) toc }.onFailure { e -> logger.error("Error generating summary for TOC entry: {}", toc.startDate, e) }.getOrNull() } }.mapNotNull { it.await() }
65. SUMMARIZATION val processed = tocList.map { toc -> async(Dispatchers.IO) { runCatching { logger.info("Generating summary for TOC entry: {}", toc.startDate) val response = summarizationChatClient .prompt() .user(toc.concatenatedUtterances.orEmpty()) .call() .chatResponse() toc.summary = response ?. result ?. output ?. text logger.info("Successfully generated summary for TOC entry: {}", toc.startDate) toc }.onFailure { e -> logger.error("Error generating summary for TOC entry: {}", toc.startDate, e) }.getOrNull() } }.mapNotNull { it.await() }
66. SUMMARIZATION val processed = tocList.map { toc -> async(Dispatchers.IO) { runCatching { logger.info("Generating summary for TOC entry: {}", toc.startDate) val response = summarizationChatClient .prompt() .user(toc.concatenatedUtterances.orEmpty()) .call() .chatResponse() toc.summary = response ?. result ?. output ?. text logger.info("Successfully generated summary for TOC entry: {}", toc.startDate) toc }.onFailure { e -> logger.error("Error generating summary for TOC entry: {}", toc.startDate, e) }.getOrNull() } }.mapNotNull { it.await() }
67. SUMMARIZATION val processed = tocList.map { toc -> async(Dispatchers.IO) { runCatching { logger.info("Generating summary for TOC entry: {}", toc.startDate) val response = summarizationChatClient .prompt() .user(toc.concatenatedUtterances.orEmpty()) .call() .chatResponse() toc.summary = response ?. result ?. output ?. text logger.info("Successfully generated summary for TOC entry: {}", toc.startDate) toc }.onFailure { e -> logger.error("Error generating summary for TOC entry: {}", toc.startDate, e) }.getOrNull() } }.mapNotNull { it.await() }
68. QUESTION EXTRACTION Related U tt erances LLM Embedding Model Extract questions Vectorize each question Store each question
69. QUESTION EXTRACTION fun updateSummary(tocData: List<TOCData>) { tocData.forEach { toc -> tocDataRepository.updateField(toc, `TOCData$`.SUMMARY, toc.summary) } } fun updateQuestions(tocData: List<TOCData>) { tocData.forEach { toc -> tocDataRepository.updateField(toc, `TOCData$`.QUESTIONS, toc.questions) } }
70. QUESTION EXTRACTION fun updateSummary(tocData: List<TOCData>) { tocData.forEach { toc - > tocDataRepository.updateField(toc, `TOCData$`.SUMMARY, toc.summary) } } fun updateQuestions(tocData: List<TOCData>) { tocData.forEach { toc - > tocDataRepository.updateField(toc, `TOCData$`.QUESTIONS, toc.questions) } } runBlocking { coroutineScope { awaitAll( async { summarizationWorkflow.run() }, async { questionGenerationWorkflow.run() } ) } }
71. QUESTION EXTRACTION @RedisHash data class Question( @Id var timestamp: String, var utterancesConcatenated: String, var utterances: List<Utterance>, @Vectorize( destination = "embeddedQuestion", embeddingType = EmbeddingType.SENTENCE, provider = EmbeddingProvider.OPENAI, openAiEmbeddingModel = OpenAiApi.EmbeddingModel.TEXT_EMBEDDING_3_LARGE ) var question: String, @VectorIndexed( algorithm = VectorAlgorithm.HNSW, dimension = 3072, distanceMetric = DistanceMetric.COSINE, ) var embeddedQuestion: ByteArray? = null )
72. TRANSLUNAR INJECTION
73. ENHANCING RESPONSES User Prompt LLM Embedding Model Vectorize Hydrate the prompt Semantic Search Final Prompt Trigger the LLM
74. fun searchByQuestion(embedding: ByteArray): List<QuestionSearchResult> { val stream: SearchStream<Question> = entityStream.of(Question :: class.java) return stream .filter(`Question$`.EMBEDDED_QUESTION.knn(3, embedding)) .sorted(`Question$`._EMBEDDED_QUESTION_SCORE) .map(Fields.of(`Question$`._THIS, `Question$`._EMBEDDED_QUESTION_SCORE)) .collect(Collectors.toList()) .map { QuestionSearchResult(it.first, it.second) } } RETRIEVE SIMILAR QUESTIONS OR SUMMARIES
75. @Bean fun ragChatClient( chatModel: ChatModel, ): ChatClient { return ChatClient.builder(chatModel) .defaultSystem(DEFAULT_PROMPT) .build() } companion object { private val DEFAULT_PROMPT = """ You are an expert assistant specializing in the Apollo missions. Your goal is to provide accurate, detailed, and concise answers to user inquiries by utilizing the provided Apollo mission data. Rely solely on the information given below and avoid introducing external information. """.trimIndent() } CONFIG A CHAT CLIENT
76. fun generateResponse(query: String, data: String): String { val response = ragChatClient .prompt() .system("Apollo mission data: $data") .user("User question: $query") .call() .chatResponse() val enhancedAnswer = response ?. result ?. output ?. text ? : "" logger.info("AI response: {}", enhancedAnswer) return enhancedAnswer } GENERATE RESPONSE
77. @Bean fun ragChatClient( chatModel: ChatModel, ): ChatClient { return ChatClient.builder(chatModel) .defaultSystem(DEFAULT_PROMPT) .build() } companion object { private val DEFAULT_PROMPT = """ You are an expert assistant specializing in the Apollo missions. Your goal is to provide accurate, detailed, and concise answers to user inquiries by utilizing the provided Apollo mission data. Rely solely on the information given below and avoid introducing external information. """.trimIndent() } WHY MULTIPLE CHAT CLIENTS?
78. @Bean public ChatClient analysisChatClient( ChatModel chatModel, ChatMemory chatMemory, DateTimeTools dateTimeTools) { return ChatClient.builder(chatModel) .defaultAdvisors(MessageChatMemoryAdvisor.builder(chatMemory).build()) .defaultTools(dateTimeTools) .defaultSystem(DEFAULT_PROMPT) .defaultOptions( ToolCallingChatOptions.builder() .internalToolExecutionEnabled(true) .build()) .build(); } SEPARATION OF CONCERNS
79. LUNAR LANDING
80. SEMANTIC CACHE
81. User Prompt Regular Pipeline (LLM) Embedding Model Vectorize Cache Hit Check semantic cache Return previously generated response Cache Miss
82. @Bean fun vectorizer(): SentenceTransformersVectorizer { return SentenceTransformersVectorizer("Xenova/all-MiniLM-L6-v2") } @Bean fun semanticCache( jedis: UnifiedJedis, vectorizer: BaseVectorizer ): SemanticCache { return SemanticCache.Builder() .name("semantic-cache") .distanceThreshold(0.2F) .ttl(360) .redisClient(jedis) .vectorizer(vectorizer) .build() } REDIS VECTOR LIBRARY
83. @Bean fun vectorizer(): SentenceTransformersVectorizer { return SentenceTransformersVectorizer("Xenova/all-MiniLM-L6-v2") } @Bean fun semanticCache( jedis: UnifiedJedis, vectorizer: BaseVectorizer ): SemanticCache { return SemanticCache.Builder() .name("semantic-cache") .distanceThreshold(0.2F) .ttl(360) .redisClient(jedis) .vectorizer(vectorizer) .build() } val cached = semanticCache.check(request.query) if (cached.isPresent) { val c = cached.get() return CachedResponse( query = request.query, answer = c.response, ) }
84. @Bean fun vectorizer(): SentenceTransformersVectorizer { return SentenceTransformersVectorizer("Xenova/all-MiniLM-L6-v2") } @Bean fun semanticCache( jedis: UnifiedJedis, vectorizer: BaseVectorizer ): SemanticCache { return SemanticCache.Builder() .name("semantic-cache") .distanceThreshold(0.2F) .ttl(360) .redisClient(jedis) .vectorizer(vectorizer) .build() } semanticCache.store(query, answer) val cached = semanticCache.check(request.query) if (cached.isPresent) { val c = cached.get() return CachedResponse( query = request.query, answer = c.response, ) }
85. EARTH’S ATMOSPHERE REENTRY
86. ⡋ 2025 Redis Ltd. All rights reserved. 86
87. - How to enhance search capabilities (Vector Similarity Search) - How to improve VS search accuracy (Chunking, Grouping, Summarization, Extraction) - How to improve LLMs responses (Retrieval Augmented Generation) - How to save speed and time when dealing with LLMs (Semantic Cache) WHAT WE SAW TODAY:
88. KOTLIN NOTEBOOKS
89. Exact vs approximate nearest neighbors in vector databases What is a vector database? What is an embedding model? Redis Redis Redis
90. REDIS AI RESOURCES REPOSITORY
91. Software Engineer (Developer Advocacy) Raphael De Lio * Slides GitHub YouTube