Track your progress.0 of 19 modules complete

Resources / AI System Design

AI System Design archive

A structured path through designing AI systems end to end. 19 modules covering requirements, data and training pipelines, serving, LLM, RAG and agent architectures, reliability, security, observability, evaluation and MLOps, with case studies and interview revision at the end.

19 modules168 sections29 diagrams

Curriculum structure

FoundationsWhat AI system design is and how to scope it

01Introduction to AI System Design7 sections 02Requirements and Use Case Analysis7 sections

Data and modelsPipelines, training and serving

03Data Pipeline and Data Engineering9 sections 04Model Training Pipeline9 sections 05Model Serving and Inference Architecture8 sections

Application architecturesLLMs, RAG, agents and the components beneath them

06LLM System Design8 sections 07RAG System Design9 sections 08Agentic AI System Design8 sections 09AI Agents Architecture8 sections 10System Components Deep Dive11 sections

Production qualityReliability, security, observability, evaluation and operations

11Reliability, Fault Tolerance and High Availability10 sections 12Security, Privacy and Responsible AI10 sections 13Monitoring, Logging and Observability9 sections 14Evaluation and Testing10 sections 15MLOps and LLMOps10 sections

In practiceCase studies and a full project

16Real-world Case Studies8 sections 17End-to-End Project Design9 sections

RevisionInterview prep, takeaways and trends

18Interview Preparation9 sections 19Key Takeaways and Future Trends9 sections

Topic threads

LLM applications

From a plain LLM app to RAG to agents.

06LLM System Design 07RAG System Design 08Agentic AI System Design 09AI Agents Architecture

Model lifecycle

Data, training, serving and the MLOps loop that ties them together.

03Data Pipeline and Data Engineering 04Model Training Pipeline 05Model Serving and Inference Architecture 15MLOps and LLMOps

Reliability & operations

Staying up, seeing inside the system and running it day to day.

11Reliability, Fault Tolerance and High Availability 13Monitoring, Logging and Observability 15MLOps and LLMOps

Evaluation

Success metrics, evaluation and testing, then evaluation inside a full project.

02Requirements and Use Case Analysis 14Evaluation and Testing 17End-to-End Project Design

Safety & security

Component-level access control and the wider security and privacy view.

10System Components Deep Dive 12Security, Privacy and Responsible AI

Applied examples

Case studies, then a worked end-to-end project.

16Real-world Case Studies 17End-to-End Project Design