Cloud Native Inference Day: 2026 is the year of Inference

Last year, the world raced to build intelligence. This year, we are racing to deploy it. At NVIDIA GTC, it was made clear, the inference inflection point has arrived. AI is no longer just about models; it’s a systems challenge defined by one question: who runs them best? Every application is becoming intelligent, powered by GPUs, accelerated computing, and cloud-native systems built for inference at scale.

And if you’ve been with us, you already know this story.

After the incredible success of KCD Delhi, where hundreds of builders, engineers, and community members came together, thank you for making it what it was, we’re not slowing down. At CNCF Delhi, we’ve always believed in pushing the community forward. And now, we’re coming back stronger, announcing our biggest meetup yet, with ~300 folks, focused entirely on Inference.

This isn’t just another meetup. It’s a continuation of the momentum we built together. We’ll make sure you understand how inference actually works, break down latency vs throughput trade-offs, and explore how Kubernetes, along with tools like vLLM and llm-d, is shaping the future of cloud-native AI. Expect a powerful lineup of speakers, real conversations, and a room full of people building what’s next.

Learn. Network. Explore.

CFPs are open now, please apply using the link: https://forms.gle/MC81J2CEwzCm4CJi7