Confluent

Over the past decade, AI has been trained on historical data. The next decade will be defined by AI systems that operate in the moment, continuously ingesting, reasoning, and acting on real-time data. That’s exactly what we saw in Cohort-3 of our Data Streaming AI Accelerator program. Tim Graczewski breaks down the key themes, architectures, and startups building this future below! ⤵️

🛒 Did you know 10% of your grocery store's items might be out of stock right now? 🤯 It is wild to think about, but even in 2026, many major retailers still rely on workers walking the aisles with clipboards or hand scanners to check inventory. The result? The industry standard for inventory accuracy is only sitting between 65% and 80%. For a store operating on razor-thin margins, an out-of-stock item doesn't just mean a lost sale—it can mean losing a customer forever. On the newest episode of Life Is But A Stream, I sat down with Kevin Johnson, CEO of Focal Systems, to talk about how they are blowing up the old retail model using AI and real-time computer vision. Instead of a daily manual scan, Focal Systems uses hundreds of cameras to process between 5 and 6 MILLION images a day. They feed all this into a real-time data stream to provide an always-on, 95% accurate picture of the store! Instead of walking aimlessly, store clerks receive highly optimized, geographically grouped tasks telling them exactly what to restock, from what backroom bin, and to what specific shelf. When you get your data flowing in real time, you don't just solve one problem—you completely revolutionize the business. 📺 Watch the full episode here: https://lnkd.in/gnPf5nZj 👇 Q: What is the most frustrating item to find out of stock when you're at the grocery store? Drop it in the comments! #DataStreaming #Confluent #Kafka #DataInMotion #Retail #EventDrivenArchitecture

🌟 Meet our Catalyst of the Week, Elad! Elad Leev is a frequent global conference speaker, including several Kafka Summits, and an active contributor to the Kafka and Flink ecosystem. He has developed numerous open-source tools and regularly shares insights on engineering, data technologies, and streaming systems through his blog. Beyond speaking and writing, Elad supports the community by advising companies adopting streaming technologies, helping startups explore new ideas in the space, and mentoring engineers transitioning into data-related roles. 👉 Learn more about our Catalyst program here: https://cnfl.io/3P50GGo

The 2026 Confluent Partner Awards APAC winners are here! We’re beyond excited to celebrate these game changers in data streaming. 🍾 Congrats to our winners: 🏆 Emerging Partner of the Year - iZeno 🏆 OEM Partner of the Year - Infosys 🏆 Emerging Partner of the Year, India - Giniminds 🏆 Co-sell Partner of the Year - iValue Group 🏆 GSI of the Year - Tata Consultancy Services 🏆 Training Partner of the Year - DataCouch 🏆 CSP Market Place Partner of the Year - Google Cloud 🏆 Co-Delivery Partner of the Year - Enhanced Software Solutions Pvt. Ltd. 🏆 Co-Marketing Partner of the Year - Amazon Web Services (AWS) 🏆 ECO System Synergy Partner of the Year - Celebal Technologies 🏆 Confluent Partner Evangelist of the Year - Sonali Samal, Giniminds 🏆 APJ Partner of the Year - Mitra Integrasi Informatika, PT

Simpler to Build On. Simpler to Run. Confluent Platform 8.2 is here! Built on Apache Kafka® 4.2, this update extends and simplifies what you can do with Kafka and Apache Flink. Build rich stateful apps with Flink SQL, natively handle task-queue workloads with Queues for Kafka, or tap into seamless client switchover for migrations, upgrades and DR. Get a closer look at the latest and greatest in Confluent Platform: https://cnfl.io/4vyUlDZ

If your AI agents hallucinate, stall in production, or deliver stale answers, you don’t have a model problem. You have a data problem. The solution? Context engineering. Download our latest ebook to learn how to master it and deliver fresh context for reliable AI: https://cnfl.io/4cRDjtZ

I will be at Google Cloud Next, speaking and hosting an event with Anthropic. If you're there and want to chat all things AI, hit me up. You can find additional information about my talk here: https://cnfl.io/4dj4ZIn

Don't miss our Context Engineering Lightning Talk next week at #GoogleCloudNext! Join Sean Falconer as he showcases how Confluent and Google Cloud work hand-in-hand to enable accurate, autonomous action without the lag you’d find in traditional data stacks. → https://cnfl.io/4dj4ZIn | Google Cloud Partners

April isn’t slowing down! If you’re looking to connect with the data streaming community, the second half of the month is stacked with meetups around the world. Here’s where we’ll be next: 📅 April 21 📍Copenhagen, Denmark | From Raw Data to Real Insight with Elastic, hosted by Nine A/S ➡️ https://cnfl.io/425G1VZ 📅 April 22 📍Athens, Greece | Apache Kafka® Meetup, hosted by foobar io Greece ➡️ https://cnfl.io/483Ddw5 📅 April 25 📍Mumbai, India | Apache Kafka® x Apache Flink® x Apache Iceberg™ Meetup, hosted by IIT Bombay | Join Mustafa Bhanpurawala, Shrinivas Vishnupurikar Kulkarni, and Ajinkya Kolhe ➡️ https://cnfl.io/4tNc7So 📍Chennai, India | Apache Kafka® Meetup, hosted by Facilio| Join Ritesh Patni ➡️ https://cnfl.io/4tRFjb4 📍Bengaluru, India | Apache Kafka® Meetup, hosted by ZopSmart | Join Faheem Mohammed, Aryan Mehrotra ➡️ https://cnfl.io/4vvSzDy 📅 April 28 📍Warsaw, Poland | From Edge to Gateway: Architecting Your OpenTelemetry Collector Deployment with Elastic ➡️ https://cnfl.io/4vvSwYo 📅 April 29 📍Madrid, Spain | Apache Kafka® x Grafana with Elastic, hosted by MasOrange | Join Dani Torramilans ➡️ https://cnfl.io/4vvSwHS 📅 April 30 📍Krakow, Poland | Data Streaming Meetup, hosted by VirtusLab ➡️ https://cnfl.io/41uX97B

In this new episode of Life Is But A Stream, Kevin Johnson joins Joseph Morais to share how Focal Systems (our first-ever Data Streaming Startup of the Year!) built a computer vision platform that processes millions of shelf images daily with Apache Flink on Confluent Cloud. From 95% inventory accuracy to real-time task automation, see what happens when AI-powered products are built on a foundation that can actually keep up: https://cnfl.io/4sCYMLw