Redis

Redis loves agents, and agents love Redis because agents are the new consumers. We’ve spent years as the fast state management layer next to the application, and in the agent era, that same layer holds agent state and the procedures agents learn over time. Redis Iris puts agent memory, a context gateway, agent playbooks, and agent mesh on top of Redis. CEO Rowan Trollope walks through how the pieces fit on Tech Talks with AIM: https://lnkd.in/evxWxDRG

If you’re working through application performance or scaling challenges, this one’s worth a spot on your Ignite agenda. Sitecore built Content Hub to help organizations manage, organize, and distribute content and digital assets across channels. At Microsoft Ignite, we’ll look at how Sitecore uses Azure Managed Redis to support Content Hub’s Azure Kubernetes Service-based workloads. Explore the role of Redis beyond caching, how it fits alongside databases and application services, and what Sitecore’s experience can teach you about building responsive, scalable applications on Azure. Join Justine Cocchi and Shruti Pathak for: Beyond the cache: How Sitecore builds to scale on Azure Managed Redis | BRK374 November 19, 2026 | 5:15–6:00 PM PT Join online or in person at Moscone Center, San Francisco. Coming to Ignite in person? Stop by the Cosmos DB DocumentDB & Redis in Azure booth. We’d love to hear what you’re building and talk through your questions with Microsoft and Redis experts. Our blog has the session link to register or add it to your schedule, plus where to find us and resources to explore before the event: https://lnkd.in/ea83MQFx Hope to see you there!

Demo: flawless ✨ Production:  😵💫😖 Production has a way of finding the edge cases. But we’re working on it 😏 Join us Nov 12: https://lnkd.in/ecqbgdzY

Your AI spend is growing. Is the return? Join Redis CEO Rowan Trollope and CFO Tony Tiscornia to explore how to turn AI investment into measurable business value.

Traffic spikes from promotions, viral products, new regions, and events like Black Friday require the data layer to scale up and back down while it keeps serving live traffic. Smooth Scaling in Redis Cloud Pro changes how data moves during that process: it relocates only the hash slots that need to move, sends them directly to their destinations, and hands each one over in a single step. Scaling requests are made the same way as before. Check out how it works; Yoav Peled explains it here: https://lnkd.in/eN6qDRSi

The real world is messy: humans don’t follow scripts, inputs can seem random, and every channel, agent communication, and handoff is another potential failure point. Join us on November 12 as we uncover the latest with Redis Iris. Be the first in the know. 🤫

A standard redis-py client connects to one configured endpoint, so handling a regional outage means every application has to build its own health monitoring, endpoint selection, failover, and failback. MultiDBClient moves that logic into the client. Paired with Active-Active Redis, which uses CRDTs to converge concurrent updates across regions, it lets Python applications fail over to another region on their own. To learn more, Vladyslav Vildanov walks through the configuration, including health-check policies, circuit breakers, and automatic failback: https://lnkd.in/ebZuEV63

Something new is coming for your agents. Register now to get the first look at our new capabilities on Nov 12: https://lnkd.in/e_TbGmcR

Production AI apps need the right context and data in real time, and we are expanding our collaboration with AWS to make that easier to deliver. Our expanded strategic collaboration agreement covers technology integration, customer migration, generative AI, AWS Marketplace, and go-to-market programs. Redis Cloud works with Amazon Bedrock for RAG, vector search, and semantic caching, and we will produce reference architectures with AWS that help teams move from prototype to production. Read more: https://lnkd.in/erZV-XPi

You all know I'm fascinated by agent memory. Well, I just dove into Redis' Agent Memory Benchmark report. They ran an agent memory test with 500 questions against a bunch of systems that support agentic memory, and they found that agents remember better when they keep original conversation excerpts AND extracted facts. Makes sense when you think about it, because the extracted facts help the agent keep track of things as they change while the excerpts contain the exact details. Combine both and the agent can better piece together what it needs. Fascinating stuff if you wanna check it out: https://fandf.co/4z3LxaL Thanks Redis for sending the report my way and partnering on this post.