Model serving & orchestration
KServe, vLLM, Ray Serve, Knative, autoscaling and GPU scheduling.
A practitioner‑first community where developers, platform engineers and data teams swap hard‑won lessons, demos and code for running AI in production.
We’re a community focused on the intersection of Cloud Native and AI — from model serving on Kubernetes to observability, security, cost control and data pipelines. We aim to be vendor‑neutral, open‑source friendly and hands‑on.
KServe, vLLM, Ray Serve, Knative, autoscaling and GPU scheduling.
Embeddings, retrieval patterns, vector databases, latency & quality trade‑offs.
Feature stores, CI/CD for models, model registries and drift monitoring.
OpenTelemetry, tracing GenAI, prompt and token cost visibility.
SBOM, supply chain security, secrets, isolation and responsible AI.
Internal developer platforms, golden paths and self‑service for AI teams.
Be kind. Be respectful. No harassment. Keep it inclusive and welcoming for everyone. Organizers may remove anyone who violates this standard.