Redis

Software is about to eat the world - for real this time. In the same way that mobile phones, editing software and YouTube made millions of creators AI will enable everyone to be a creator of software. Just like videos there will be endless streams of software / app slop. But the coming wave will change the world of software as we know it. In my own company we have internal software being created by pre-sales reps, by marketing folks, by sellers, by FP&A, by executive admins, by real estate folks and more. We are making everything at redis for that future world. Memory for agents. State for agents. Optimizing context windows to reduce token spend. Delivering data to agents. Caching data for agents.

The future of agentic AI won't be defined by better models, but by better context. @simba_khadder joined @alexborek on the Data Masterclass Podcast to break down why context engineering will decide who wins with AI in 2026. Full episode: youtube.com/watch?v=jqmxre…

Q: What is agent memory, and why does it matter? A: “Agent memory helps an AI application remember useful context across a conversation or across sessions. Memory helps reduce that friction by helping agents remember which tool calls they used before that actually worked and how to structure the call better so it ultimately achieves what the user wanted…Used well, memory makes agents feel more continuous, personalized, and helpful.”

Q: Why do AI agents still give generic answers when we already have RAG? A: “Because basic RAG often gives the model only one slice of the picture… That is why teams are moving from simple document retrieval to richer context strategies that combine documents, structured data, memory, APIs, and real-time state.”

Context engineering FAQ from @Simba_Khadder. Why do agents still give generic answers, RAG vs. agentic RAG, memory vs. semantic caching, build vs. buy — a thread on the questions teams are actually asking right now. 🧵 redis.io/blog/faq-real-…

Watch the tutorial here: youtu.be/qhsVMiBjxM0?si…

First time speaking at @WeAreDevs conference in Berlin! Gave a talk on effective context management being a moat and strategies to think about when working on agent memory systems! Incredible work by the @Redisinc team at the booth!

Markdown is agent memory for one person. It breaks the moment a second person shows up. @cole_medin builds the version that scales: a support agent on Redis Iris, where one tool call pulls every delayed order without any schema digging or text-to-SQL. Watch Cole build it end to end, then build your own: youtube.com/watch?v=R-5_2n…

Additionally, Redis Data Integration is now GA in Redis Cloud on AWS. Get started with RDI today wherever you build: redis.io/blog/redis-dat…

$800K of revenue loss saved. Stability for 10 million daily users. Different problems, both solved by the same fix: getting rid of the lag between where data lives and where it's actually used with Redis. Redis Data Integration continuously syncs data from existing databases and warehouses into Redis in near real-time, from initial load to every subsequent change. Connect your data and get started: redis.io/data-integrati…

It comes down to the agent being stateful, making decisions based on what it's already done, not just what it knows about you. ➡️ A coding agent that remembers which file or module to touch to ship a new feature ➡️ A support agent that recognizes a problem it's seen before and remembers the ticket it already filed Personalization is powerful on its own. But this is the difference between an agent that personalizes and one that also compounds: it learns from its own mistakes and gets better run over run. The way this typically works under the hood: as the agent works, important information gets extracted from the conversation and stored outside the model as long-term memory. When it's needed again, it's retrieved by semantic similarity plus metadata filters, while the session itself gets summarized and trimmed instead of replaying the full history every time. @Redisinc has something called Redis Iris, their real-time context engine for agents. Agent Memory is one of the pieces inside it: it manages both short-term session state and long-term memory that carries across tasks, so an agent has a persistent record to work from instead of starting from zero every run. Memory that compounds is what turns an agent from a tool you use into one that gets better because you used it. 📖 If you want to go deeper on this, Simba Khadder wrote a great FAQ covering Agent Memory and the rest of Redis Iris, their context engine:
redis.io/blog/faq-real-…

When people hear "agent memory," what commonly comes to mind is personalization.... but there's another piece that gets less attention. Let's take a walk with @simba_khadder, who leads Context Engine at @Redisinc, as he explains ✨

Context is what separates a useful agent from a frustrating one. Redis Agent Memory gives agents short-term memory within a session and durable memory across conversations. Most agents lose context the moment a session ends. Redis Agent Memory solves that with a managed memory layer that stores session-scoped working memory, extracts high-signal facts and decisions from each conversation, and keeps them searchable so the agent can pull what's relevant when it matters. More on Redis Agent Memory: redis.io/agent-memory/

Hey Go developers, I have great news for you. You now have a version of RedisVL available for @golang. This project enables you to develop AI apps that use @Redisinc. It supports all the great features pioneered by the original RedisVL for Python. github.com/redis-develope…

Agents use 4x more tokens than chat. Redis LangCache intercepts the redundant calls and returns cached responses instantly, saving up to 90% on API costs. Instead of calling an LLM for every request, LangCache checks if a similar response has already been made and returns it from cache in milliseconds. Fully managed semantic caching via a REST API—nothing to build and no redundant calls to pay for. See how much Redis LangCache can save: redis.io/langcache/

Karthik Kishore, Regional Director, @Redisinc shares his perspective at #ETCIOAC26 on how enterprises can rethink user experience in a world driven by apps and AI. Know More - lnkd.in/dnSjZhB #ETCIOAC26 #ETCIO #Redis #AI #EnterpriseAI #DigitalTransformation

Agents are only as good as the context they run on. Redis Iris is a unified, real-time context engine that delivers fresh, relevant context so agents perform at scale. Context Retriever navigates business data. RDI keeps it fresh. Agent Memory makes it persist. Redis Search retrieves it fast. LangCache eliminates the redundant calls. Together, that's Redis Iris. Get started with Redis Iris for free: #redis-iris" target="_blank" rel="nofollow noopener">redis.io/iris/#redis-ir…

Redis metrics have always told you what the server is doing. Now they tell you what your app is doing. Native OTel support is live in redis-py, go-redis, and node-redis. The client layer is finally visible. Read more: redis.io/blog/native-op…

There's now a @Redisinc official SDK for @FastAPI 🚀 We had some conversations, I suggested some API design ideas and (future-proof) constraints on the FastAPI side, @tishun did an amazing job implementing everything 🙌 github.com/redis/fastapi-…