Synthora
Sign inGet started
IntermediateSelf-paced (~20 hrs) Synthora STA certified HRDF-claimable

AI Agents & Agentic Workflows — Build Autonomous AI Systems

Design, build and deploy autonomous AI agents using LangChain, CrewAI and the Anthropic MCP.

About this course

AI agents are the defining technology of 2026 — Gartner projects 40% of enterprise applications will embed task-specific agents by end of year. This hands-on course teaches you to build multi-agent systems that plan, use tools, remember context and execute complex workflows autonomously. Covers LangChain, LangGraph, CrewAI, the Anthropic Model Context Protocol, OpenAI Assistants, and production deployment patterns.

What you'll achieve

  • Design single and multi-agent architectures for real business problems
  • Build agents with tool use, memory, planning and reflection capabilities
  • Implement RAG-powered agents with vector database integration
  • Orchestrate multi-agent teams with CrewAI and LangGraph
  • Deploy agents to production with monitoring and guardrails
  • Apply the Anthropic Model Context Protocol (MCP) for tool integration

Curriculum

  1. Module 1

    AI Agents Fundamentals

    What are agents? · ReAct pattern · Tool use · Memory types · Planning strategies

  2. Module 2

    LangChain & LangGraph Deep Dive

    Chains vs agents · LangGraph state machines · Conditional edges · Human-in-the-loop

  3. Module 3

    Tool Integration & Function Calling

    OpenAI function calling · Custom tools · Web search · Code execution · API tools

  4. Module 4

    Memory & Context Management

    Short-term memory · Long-term memory · Vector stores · Episodic memory · Semantic search

  5. Module 5

    Multi-Agent Systems with CrewAI

    Roles & personas · Task delegation · Agent collaboration · Crew orchestration

  6. Module 6

    Anthropic MCP & Claude Integration

    Model Context Protocol · MCP servers · Claude tool use · Building MCP tools

  7. Module 7

    Production Deployment & Monitoring

    LangSmith observability · Guardrails · Cost management · Error handling · Scaling patterns

  8. Module 8

    Capstone: Enterprise Agent System

    Problem scoping · Architecture design · Build & test · Demo & documentation

Who this is for

  • Software developers & backend engineers
  • Data scientists moving into AI engineering
  • Product managers building AI-powered features
  • Enterprise teams automating workflows with AI

Tools & technologies

LangChain LangGraph CrewAI Claude API OpenAI API Pinecone FastAPI Docker

Prerequisites

  • Python basics (functions, classes, loops)
  • Familiarity with APIs and REST concepts
  • Basic understanding of LLMs (ChatGPT-level)