openshift-ai

Data science pipelines and model serving using Red Hat OpenShift Data Science

Production-pattern Red Hat OpenShift AI 3.4.0 platform with bare-metal ESXi, GPU passthrough, KServe RawDeployment, DeepSeek R1 inference at 12–17 tok/s

Deploy Cisco AI Pods with Ansible and Python

This guide is designed to help you deploy OpenShift on bare metal for a variety of workloads, providing flexibility, scalability, and advanced configurations tailored to diverse use cases.

Sanitized reference architecture for building a private family-focused AI platform on Red Hat OpenShift, OpenShift AI, OpenShift Virtualization, AAP, Strix Halo, and Apple Silicon.

Validated Pattern that deploys OpenShift AI and OpenShift Pipelines.

MLOps and compute acceleration with Openshift AI

OpenShift 4.20 installation (Day 0/1/2) & architecture matrix across bare metal, hypervisors, cloud & air-gap. AI-generated with Gemini 3.8 Flash; not yet tested on live clusters, serves as an architecture reference.

Red Hat OpenShift AI Model Registry Deployment and Automated Model Ingestion

This repository contains the code for a workshop on how to use different tools to monitor and manage an edge-anomaly-detection application.

4 production-ready Agent Skills for OpenShift AI 3.4 — platform architecture, MaaS governance, AI gateway routing, and inference optimization

Personal blog exploring AI infrastructure, platform engineering, and cloud architecture. Topics include NVIDIA AI Infrastructure & Operations, InfiniBand networking, Kubernetes/OpenShift, agentic AI with LangGraph, and enterprise MLOps.

Edge AI inference on OpenShift AI — GGUF LLM serving via llama.cpp and Lemonade, YOLO26 object detection and segmentation, and a unified multi-model web UI, running on a three-node SNUC cluster with AMD Ryzen AI iGPUs.

Production-ready AgentOps showcase on OpenShift AI: secure agent isolation (OpenShell), guardrails (NeMo), observability (MLflow), and a live BYOA demo with OpenClaw.

OpenShift Validated Reference Design for Enterprise AI

Charter, guidelines, and plugin catalog for the Red Hat AI Community Plugins initiative

Not Every Question Needs the Whole GPU: BF16, FP8, INT4 and speculative decoding on Llama 3.1 70B and Qwen3.8-27B, measured on H200s and MIG slices. PyTorch Conference NA 2026 demo.

My Personal Blog, learn, share and discover new things

AWS 에 OpenShift 를 IPI 로 설치하고 agent OSS 스택과 Red Hat OpenShift AI 를 올려 GPU 서빙·LoRA 튜닝까지 해 본 뒤 깨끗하게 지우는 실습 레포

Making Inference work for you with vLLM and RHOAI Enterprise Model Serving