Redis

Ever wonder how AI apps make sense of images, text, or video? It starts with a vector database. From vectorization to HNSW indexing, Brian Sam-Bodden, Redis’ Principal Applied AI Engineer, reveals how they work, why they matter, and how techniques like cosine similarity, HNSW indexing, and vectorization power smarter search and recommendations. Dive in: https://lnkd.in/epKu3unB

AI’s impact isn’t just about its capabilities. It also comes from the people who bring AI to life. Our VP of AI Product Management, Manvinder Singh, recently shared practical advice on turning AI potential into real-world results alongside other tech leaders. Read the full article for more: https://lnkd.in/emF7BAGT

We’re co-hosting a Founder Game Night next Thursday, July 24, in Brooklyn with Tavily, Eden, and Jake Z. at Fractal Tech. Come for an evening of board, card, and social games. Stay for the great conversations with other folks building in AI infrastructure. Bring your A-game. We’ll bring the drinks and snacks. Register here: https://lnkd.in/ew-gJNae

Seriously. He even mentioned changing our name to rAdIs. Ok, that's not true, but he really does want you to go here: https://lnkd.in/gPzeK6qn

Retrieval Augmented Generation (RAG) is one of the most powerful architectural patterns in GenAI today—combining the strengths of LLMs with real-time, external context from your own data. Learn more about the how and why of RAG with Brian Sam-Bodden, Principal Applied AI Engineer at Redis, including: ▶️ Query rewriting, dense retrieval, and semantic chunking ▶️ How to structure your data for better grounding ▶️ What’s happening behind the scenes ▶️ Why RAG improves accuracy, reduces hallucinations & keeps outputs fresh Check out the full video for more: https://bit.ly/3IwN2sp

If you're working with vector search, LLM pipelines, or unstructured data, you’re relying on embedding models. Redis Developer Advocate Raphael De Lio breaks down: ✅ What embedding models are and how they represent meaning using vectors ✅ How they help computers understand text, images, and unstructured data ✅ Why embeddings drive use cases like search, recommendations, and fraud detection ✅ Where to find ready-to-use models and how to get started quickly Watch the full video: https://bit.ly/4nJn4lP

Traditional search matches words. Semantic search matches meaning. Visual search matches images based on visual similarity. In this video, I explain how visual search uses image embeddings and vector similarity search to understand what an image represents, not just how it looks at the pixel level — going far beyond raw pixel comparison. https://lnkd.in/e9VP2ENh

When 100M+ fans stream live Indian Premier League matches on JioCinema, performance can’t falter. That’s why they use Redis to power real-time features like leaderboards, live view tracking, and interactive stickers, all with sub-millisecond latency. See how it works: https://lnkd.in/gUUt6yFt

Join us in San Francisco on Sept. 4 for Redis Released, a one-day conference for teams building fast, scalable AI infrastructure. You’ll work shoulder-to-shoulder with Redis engineers, alongside experts from leading AI companies, to learn how the fastest teams are building and scaling AI systems today and preparing for what’s coming tomorrow. 🔍 Here’s a session preview worth showing up for: Ricardo Ferreira, our Developer Advocacy Lead, will share real-world strategies for keeping vector embeddings fresh in production. You’ll learn: ▶️ How to implement real-time vector synchronization at scale ▶️ Change detection strategies to avoid unnecessary reprocessing ▶️ Event-driven design patterns that keep your AI features consistent and fresh ▶️ Resilient architectural approaches that decouple vectors from shifting data models ▶️ Practical advice for developers, team leads, and architects moving from AI theory to execution Come build with us. Register here: https://bit.ly/4ktRF41

Redis Insight on Cloud lets you manage and query your Redis data directly from your browser. New query autocompletion in the web and desktop apps suggests schema, index, and key names in real time to speed up your workflow. Now available in public preview: https://bit.ly/4lP9rzQ