Liên hệ với chúng tôi
 Thời gian kéo dài 14 Giờ học

Đề cương khóa học

Introduction

  • Workflows versus agents
  • Architecture overview and typical use cases

Claude API fundamentals

  • Messages API, models and parameters
  • System prompts and structured output
  • Streaming and error handling
  • Lab: first API integration

Tool use

  • Defining tools and schemas
  • The tool-use loop and error recovery
  • Lab: an agent that calls company APIs

MCP and Agent Skills

  • MCP servers and clients
  • Using existing servers and writing a small one
  • Agent Skills for reusable expertise
  • Lab: connect an agent to a data source

Claude Agent SDK

  • Agent loop, sessions and context management
  • Permissions, hooks and subagents
  • Lab: build a research and reporting agent

Cost, quality and safety

  • Prompt caching and token budgeting
  • Guardrails and human approval steps
  • Evaluations with a small test set
  • Logging and monitoring

From prototype to production

  • Deployment options and secrets handling
  • Maintenance and versioning
  • Roadmap for the participants’ own use case

Yêu cầu

  • Programming experience in Python or TypeScript
  • Experience calling REST APIs and handling JSON
  • Basic understanding of large language models

Audience

  • Software developers and data engineers who build AI features into products and internal tools
  • Solution architects evaluating agent architectures
  • Technical teams that want to move from prototype to production

Số người tham gia


Giá cho mỗi học viên

Đánh giá (3)

Các khóa học sắp tới

Các danh mục liên quan