AI Platform Engineering Handbook
AI Platform Engineering Handbook

Production-grade learning path

Engineer the platforms that power AI at scale.

A practical handbook for moving from cloud-native foundations to MLOps, RAG, agentic systems, AI security, and production infrastructure.

Start with DevOps Explore the roadmap
14
published weeks
45
deep-dive modules
1
connected path
YOUR LEARNING PATH
01DevOpsBuild the base
02MLOpsShip models
03RAG + AgentsBuild intelligence
04Enterprise AIOperate at scale
Curriculum online • Start anywhere

A connected curriculum

From reliable delivery to intelligent systems.

The handbook follows the work of a modern platform engineer: establish secure delivery foundations, operationalize machine learning, build knowledge-aware applications, then run trusted AI services in production.

Browse all documentation

The roadmap

Build capability in the right order.

Each stage connects core engineering practices to the systems you will design and operate next.

01 — 04Foundation

Cloud-native engineering

Master dependable delivery, container security, Kubernetes operations, MLOps workflows, IaC, GitOps, and internal platforms.

  • Advanced Git & CI/CD
  • Docker, Kubernetes & observability
  • Terraform, ArgoCD & developer platforms
Explore Weeks 1–4
05 — 06Intelligence

Generative AI & RAG

Learn the building blocks behind grounded AI applications, from model APIs and prompting to retrieval and evaluation.

  • Transformers, embeddings & prompt systems
  • Vector databases & hybrid retrieval
  • Advanced RAG, citations & evaluation
Explore Weeks 5–6
07 — 10Autonomy

Agents & LLMOps

Design agentic workflows with memory, tools, MCP, observability, governance, and enterprise architecture.

  • Agent patterns & leading frameworks
  • Memory systems, tool calling & MCP
  • LLMOps, guardrails & multi-agent systems
Explore Weeks 7–10
11 — 15Scale

Trusted production AI

Take AI systems to production with security, GPU infrastructure, optimized serving, fine-tuning, and product delivery.

  • AI security, red teaming & governance
  • GPU systems, vLLM & distributed serving
  • PEFT, alignment & AI product engineering
Explore Weeks 11–15

What you will practice

Work with the systems behind useful AI.

This is not a collection of isolated tools. The material connects architecture, operations, security, and product decisions into an engineering discipline.

See the platform architecture track
01

Deploy models with confidence

Version, package, test, scale, and monitor ML inference services.

 02

Ground AI in knowledge

Use retrieval, re-ranking, citations, and evaluation to earn trust.

 03

Build agents with real tools

Connect memory, structured outputs, tool calling, and MCP safely.

 04

Run performant AI infrastructure

Understand GPU capacity, continuous batching, KV cache, and serving engines.

Start a deep dive

Choose your next system.

View all modules
Platform delivery

GitOps & Continuous Delivery

Design declarative, self-healing delivery systems for modern platforms.

 Reliable AI

AI Observability & Monitoring

Trace, measure, diagnose, and improve systems after they reach users.

 Secure by design

AI Security Fundamentals

Understand the LLM attack surface, prompt injection, and AI risk management.

H A R S H H A A's GitHub avatar

About the author

Creator of the AI Platform Engineering Handbook

H A R S H H A A

This handbook is built to help engineers connect dependable DevOps foundations with modern MLOps, platform engineering, and production AI systems—one practical capability at a time.

DevOpsMLOpsAI Platforms
Follow @NotHarshhaa on GitHub

Build your path

Start with the foundations.
Grow into AI platform engineering.

Begin Week 1