Get in Touch
 Duration 14 hours

Course Outline

Introduction

  • Distinguishing workflows from agents
  • High-level architecture and common use cases

Core Claude API Concepts

  • Messages API, available models, and key parameters
  • Crafting system prompts and managing structured output
  • Implementing streaming and robust error handling
  • Lab: Initial API integration

Implementing Tool Use

  • Creating tool definitions and schemas
  • Managing the tool-use loop and error recovery strategies
  • Lab: Developing an agent that interacts with internal APIs

MCP and Agent Skills

  • Understanding MCP servers and clients
  • Integrating existing servers and developing custom ones
  • Applying Agent Skills for reusable expertise
  • Lab: Connecting an agent to a data source

Claude Agent SDK

  • Managing the agent loop, sessions, and context
  • Configuring permissions, hooks, and subagents
  • Lab: Building a research and reporting agent

Cost, Quality, and Safety

  • Utilizing prompt caching and managing token budgets
  • Establishing guardrails and human approval mechanisms
  • Conducting evaluations with a limited test set
  • Implementing logging and monitoring solutions

Moving from Prototype to Production

  • Exploring deployment options and secret management
  • Strategies for maintenance and version control
  • Developing a roadmap for participants' specific use cases

Requirements

  • Familiarity with Python or TypeScript programming
  • Experience with REST API calls and JSON processing
  • Fundamental knowledge of large language models

Target Audience

  • Software developers and data engineers integrating AI capabilities into products and internal tools
  • Solution architects assessing agent-based architectures
  • Technical teams aiming to transition projects from prototype stage to production

Number of participants


Price per participant

Testimonials (3)

Upcoming Courses

Related Categories