Skip to main content

Microservices and System Architecture for Applied AI

Imagine microservices as modular spacecraft, and Docker as your personal spaceport for assembling them. Here you will learn how to design "orbital stations" for AI Agents: breaking down monolithic systems into autonomous satellite services, configuring their interaction through interstellar protocols (REST/gRPC), and automating deployment with CI/CD launch vehicles. These skills will allow your neural networks to scale like a galactic empire, update without downtime, and survive failures of individual components without collapsing the entire system.

Ask AI Instructions
instruction

Since these topics do not change over time, it is best to study them with a personal tutor - ChatGPT.

The learning process should be as follows:

  • you create a system prompt for ChatGPT (templates), where you describe your background, preferences, level of detail of explanations, etc.
  • copy the topic from the list (triple click), and ask ChatGPT to explain this topic to you
  • if you want to delve deeper, ask clarifying questions

At the moment, this is the most convenient way to learn the basics. In addition to concepts, you can study additional materials in the Gold, Silver, Extra sections.

  1. Gold - should definitely be studied before communicating with ChatGPT
  2. Ask AI - ask questions on each unfamiliar topic
  3. Silver - secondary materials
  4. Extra - in-depth topics

Golden​

1. Videos​

sys des: docker

2. Architectures for GenAI​

Standard Architecture for an AI Agent​

  • we need a proxy to the LLM to comply with rate-limiting
  • we need a proxy to the external API to comply with rate-limiting and caching results
  • we need a gateway in front of the backend to classify requests, determine user roles (paid tier, free tier), set limits on context size, etc., set rate-limiting, etc.

Ask AI​

Ask ChatGPT

  1. Basic Concepts of Microservice Architecture (Starter Guide)

  2. Microservices vs Monolith: a complete comparison of architectures

  3. Domain-Driven Design: basic principles for microservices

  4. API Gateway: patterns for beginners (Overview)

  5. Docker for AI services: minimally necessary practices

  6. Kubernetes: orchestration basics for AI developers

  7. Load balancing of GPU tasks: basic approaches

  8. Versioning AI models: semantic versioning

  9. A/B testing of models: production use-cases (Brief overview)

  10. Security of AI services: OAuth2/JWT in practice

  11. Rate limiting for AI operations: basics for beginners

  12. CI/CD for AI: minimal working pipeline

  13. GRPC vs REST: comparison for AI API (Concept)

  14. Caching ML model results: basic strategies

  15. Circuit Breaker: operating principle and implementation

  16. Monitoring neural network services: key metrics

  17. Caching: TTL vs invalidation (Comparative analysis)

    Caching: strategies and cache invalidation

  18. Message Queue: basic concepts and use-cases (Overview)

Silver​

System Design Tutorial, 53 min, fully with animations

Extra​