{"repo":"ld-singh/ai-factory-ops-lab","free":true,"listed":false,"github":"https://github.com/ld-singh/ai-factory-ops-lab","clone":"git clone https://github.com/ld-singh/ai-factory-ops-lab.git","description":"Hands-on GPU/HPC infrastructure operations: K8s GPU scheduling, HAMi sharing, Slurm, observability & vLLM inference. Learn it free on a laptop; validate on one cheap GPU.","language":"Shell","stars":15,"topics":["ai-infrastructure","dcgm","devops","gpu","gpu-scheduling","gpu-sharing","grafana","hami","hpc","inference"],"license":"MIT","category":"analytics","readme_excerpt":"AI Factory Operations Lab Hands-on GPU/HPC infrastructure operations — learn it free on your laptop, prove it on one cheap GPU. 📖 Read the course online → ⭐ Star this repo if it helps you — it makes the work visible and helps other engineers find it. --- A guided, learn-by-doing course in AI/HPC infrastructure operations : NVIDIA GPU infrastructure concepts, Kubernetes GPU scheduling, Slurm GPU workload management, GPU observability, inference serving, and BCM-style cluster lifecycle patterns. You don't read this repo - you run it. Each lesson has you stand something up, break it on purpose, diagnose it the way you would on a real cluster, and capture the evidence. Most of the course needs no GPU at all ; one lesson uses a single rented GPU VM and is clearly marked. Read this first - how the course works The best part: you learn most of this course on your laptop, for free. The operational discipline of running the NVIDIA GPU stack - GPU scheduling, queueing, sharing decisions, observability, Slurm, cluster lifecycle - is all taught on simulations that reproduce the real control-plane behaviour, no GPU and no cloud bill required. That's the majority of the skills, and you can run every bit of it today. When you want to feel real hardware, the real-GPU lessons put those same skills on an actual NVIDIA card - the driver-to-pod path, enforced GPU sharing, real telemetry, real benchmarks - for a budget of $5-10 : you rent a GPU VM for an hour or two and tear it down when you're ","default_branch":null,"files":null,"tree":[],"storefront":"/r/ld-singh","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/ld-singh/ai-factory-ops-lab/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}