# Introduction ### The practical handbook for infrastructure engineers > “You don’t need to be a data scientist to work with AI — but you do need to understand how it runs, scales, and is observed.” ![AI for Infra Pros](https://907582225-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FYKsakaXMV3jPqzN0DHCB%2Fuploads%2Fgit-blob-73ce1243badc5e71350cdc3b5c9d6a153288783a%2Fai4infrapros.png?alt=media) ### About this project As someone who spent years in infrastructure before transitioning into AI, I experienced firsthand how challenging it can be to bridge the gap between systems, compute, and machine learning. This repository **AI for Infra Pros**, was born from that journey. It documents the exact steps, labs, and concepts I wish I had when starting out, helping other infrastructure engineers confidently navigate their AI learning path. Each chapter connects real infrastructure knowledge to AI foundations, tools, and workloads on Azure and beyond. ## Overview **AI for Infra Pros** is a practical and technical handbook designed to help infrastructure, cloud, and DevOps professionals understand and apply **Artificial Intelligence in a secure, efficient, and scalable way**. This repository combines foundational knowledge, best practices, and real-world examples to turn infrastructure expertise into a competitive advantage in the AI era. ## What you’ll find * AI fundamentals explained from an infrastructure perspective * Architecture and automation models using **Bicep**, **Terraform**, and **YAML** * Hands-on mini-labs with **AKS**, **GPU VMs**, and **Azure Machine Learning** * Strategies for **monitoring, security, and resilience** in AI workloads * Visual glossary comparing key **Infrastructure vs. AI** concepts * An **AI Adoption Framework** tailored for technical teams ## Chapter list Below is a quick list of all chapters included in this handbook. Each chapter is self-contained and can be read independently: * Chapter 1 – AI fundamentals: [link](https://www.ai4infra.com/main-chapters/01-introduction) * Chapter 2 – Data: The fuel of AI: [link](https://www.ai4infra.com/main-chapters/02-data) * Chapter 3 – Infrastructure and compute for AI: [link](https://www.ai4infra.com/main-chapters/03-compute) * Chapter 4 – IaC and automation: [link](https://www.ai4infra.com/main-chapters/04-iac) * Chapter 5 – Monitoring and observability: [link](https://www.ai4infra.com/main-chapters/05-monitoring) * Chapter 6 – Security in AI environments: [link](https://www.ai4infra.com/main-chapters/06-security) * Chapter 7 – AI use cases for infrastructure engineers: [link](https://www.ai4infra.com/main-chapters/07-use-cases) * Chapter 8 – AI adoption framework: [link](https://www.ai4infra.com/main-chapters/08-adoption-framework) * Chapter 9 – Azure OpenAI: TPM, RPM, and PTU: [link](https://www.ai4infra.com/main-chapters/09-azure-openai-tpm-ptu) * Chapter 10 – Visual glossary: [link](https://www.ai4infra.com/main-chapters/10-visual-glossary) ## How to navigate Each chapter is designed to be **independent and complementary**.\ If you want to start quickly: | Goal | Where to start | | -------------------------------------------- | ------------------------------------------------------------------------------------------------ | | Understand how AI connects to Infrastructure | [Chapter 1 - AI fundamentals](https://www.ai4infra.com/main-chapters/01-introduction) | | Build an AI environment | [Chapter 4 - IaC and automation](https://www.ai4infra.com/main-chapters/04-iac) | | Measure and observe AI workloads | [Chapter 5 - Monitoring and observability](https://www.ai4infra.com/main-chapters/05-monitoring) | | Ensure security and resilience | [Chapter 6 - Security in AI environments](https://www.ai4infra.com/main-chapters/06-security) | | Get hands-on experience | [Mini-labs](https://www.ai4infra.com/hands-on-labs/labs) | | Translate terms and concepts | [Visual glossary](https://www.ai4infra.com/main-chapters/10-visual-glossary) | *** ## Repository structure ``` ai-for-infra-pros/ ├── docs/ │ ├── chapters/ │ │ ├── 01-introduction.md │ │ ├── 02-data.md │ │ ├── 03-compute.md │ │ ├── 04-iac.md │ │ ├── 05-monitoring.md │ │ ├── 06-security.md │ │ ├── 07-use-cases.md │ │ ├── 08-adoption-framework.md │ │ ├── 09-azure-openai-tpm-ptu.md │ │ └── 10-visual-glossary.md │ ├── extras/ │ │ ├── labs/ │ │ │ ├── bicep-vm-gpu/ │ │ │ │ └── README.md │ │ │ ├── terraform-aks-gpu/ │ │ │ │ └── README.md │ │ │ ├── yaml-inference-api/ │ │ │ │ └── README.md │ │ ├── case-studies.md │ │ ├── cheatsheets.md │ │ └── technical-faq.md │ └── images/ │ ├── ai4infrapros.png │ ├── aiforinfracon.png │ ├── designing-ha.png │ ├── example-architecture.png │ ├── framework-structure.png │ ├── full-training-pipeline.png │ ├── infrastructure-flow.png │ ├── model-life-cycle.png │ ├── predictive-analysis.png │ ├── relationship-tpm-qps-cost.png │ ├── simple-training-pipeline.png │ ├── typical-architecture-ptu.png │ └── where-to-run.png ├── README.md └── SUMMARY.md ``` ## Target audience Professionals in: * Infrastructure and Cloud (Azure, AWS, GCP) * DevOps and SRE * Solutions Architecture * Security and Governance * Data Engineering professionals who want to understand the infrastructure side of AI ## Mission To turn infrastructure knowledge into an advantage in the era of Artificial Intelligence.\ To show that **you don’t need to be a data scientist to work with AI — but you do need to understand how it runs, scales, and is observed.** ## Extra resources * [Infrastructure mini-labs for AI](https://www.ai4infra.com/hands-on-labs/labs) * [Technical Case Studies](https://www.ai4infra.com/extras-and-reference-material/case-studies) * [Cheatsheets](https://www.ai4infra.com/extras-and-reference-material/cheatsheets) * [Technical FAQ](https://www.ai4infra.com/extras-and-reference-material/technical-faq) ## Credits Created by **Ricardo Martins**\ 📍 Principal Solutions Engineer @ Microsoft\ 📖 Author of [*Azure Governance Made Simple*](https://book.azgovernance.com/) and [*Linux Hackathon*](https://linuxhackathon.com/)\ 🌐 [rmmartins.com](https://rmmartins.com) > *“AI needs infrastructure. And infrastructure needs to understand AI.”*