In Context

AI system design · for interviews

In Context

Ten AI systems, designed end to end

Cover of In Context: Ten AI systems, designed end to end, by Didarbay Nuraliyev

Contents

Before you begin

  1. Introduction
  2. How to use this book
  3. The ten systems at a glance
  4. The framework on one page
  5. The pattern map

Part I · AI building blocks

  1. 01LLM Inference PlatformServing large language models on a fleet of GPUs
  2. 02LLM GatewayOne door to every model, with budgets and failover
  3. 03Vector Search EngineFinding the nearest of billions of embeddings
  4. 04LLM Evaluation & ObservabilityKnowing whether an AI feature is good, and keeping it so
  5. 05Guardrails & Safety LayerChecking what goes into and comes out of every model

Part II · AI products

  1. 06AI Chat AssistantA ChatGPT-style assistant for tens of millions of people
  2. 07RAG Knowledge AssistantAnswers from company documents, with sources
  3. 08AI Coding AssistantCode suggestions in the editor, as fast as you type
  4. 09AI Agent PlatformRunning agents that call tools and take real actions
  5. 10Recommendation SystemChoosing what each person sees next, from millions of items

After

  1. Conclusion: ten systems, one way of thinking
  2. Appendix A: the book in ten sentences
  3. Appendix B: numbers worth knowing
  4. Further reading