₹0 for students · Certificate · Intermediate

AI Agents for Beginners

18 lessons: design, build, secure and deploy AI agents

Start free in the app

18 lessons · ~13 hours · 30-question final exam · pass mark 70%

About the course

What you’ll learn

Learn to build AI agents that plan, use tools and work together: agent frameworks, design patterns, tool use, agentic RAG, trustworthy agents, planning, multi-agent systems, metacognition, production, protocols such as MCP and A2A, context engineering, memory, browser-use agents, scalable and local deployment, and security with signed audit trails. Each lesson has a study guide, the full lesson with code samples, and a quiz. Pass every quiz and the final exam to earn a verifiable certificate.

AI agentsTool useAgentic RAGMulti-agent systemsPlanningMCPA2AContext engineeringAgent memoryAgent security

Part of the career track: GenAI and LLM Builder

How the certificate works

  1. Read each lesson’s study guide, then the full lesson and any notebooks or exercises.
  2. Pass every lesson quiz with 70% or more.
  3. Pass the final exam: 30 questions in 45 minutes.
  4. Get a certificate with a public verification link, and add it to LinkedIn.

Syllabus

18 lessons in 4 parts

Agent foundations

1
Introduction to AI Agents and Use Cases
40 min · An agent lets an LLM act through tools and knowledge.
2
Exploring AI Agent Frameworks
45 min · Frameworks provide models, tools, state, memory, orchestration and observability.

Designing capable agents

3
Agentic Design Principles
35 min · Design agents around people: space, time and core dimensions.
4
Tool Use Design Pattern
45 min · Tools let agents fetch information and take actions.
5
Agentic RAG
45 min · Agentic RAG lets the agent plan, retrieve, evaluate and iterate.
6
Building Trustworthy AI Agents
45 min · A system message framework generates consistent, detailed system prompts from a meta prompt.
7
Planning Design Pattern
40 min · Define a clear goal and decompose it into subtasks.
8
Multi-Agent Design Patterns
45 min · Use multiple agents for large workloads, complex tasks or diverse expertise.
9
Metacognition in AI Agents
40 min · Metacognition: agents reflecting on their own reasoning and actions.

Production, protocols, context and memory

10
AI Agents in Production: Observability and Evaluation
45 min · Traces capture whole agent runs; spans capture steps with timings, inputs, outputs and costs.
11
Agentic Protocols: MCP, A2A and NLWeb
40 min · MCP standardises connections between AI apps and tools/data.
12
Context Engineering for AI Agents
40 min · Context engineering manages what goes into the model's context at each step.
13
Memory for AI Agents
40 min · Memory types: working, short-term, long-term, persona, episodic, entity, structured RAG.

Advanced agent engineering

14
Exploring the Microsoft Agent Framework
45 min · Microsoft Agent Framework unifies Semantic Kernel foundations and AutoGen multi-agent ideas.
15
Building Computer Use Agents
40 min · Computer use agents operate UIs by seeing the screen and clicking, typing and scrolling.
16
Deploying Scalable Agents
45 min · Production needs identity, secrets, reliability, monitoring, cost control and safe updates.
17
Creating Local AI Agents
40 min · Local agents keep data on-device, work offline, cut costs and latency.
18
Securing AI Agents with Cryptographic Receipts
40 min · Ordinary logs can be edited or forged; receipts provide tamper-evident records.

Course content is adapted from Microsoft's "AI Agents for Beginners" curriculum (https://github.com/microsoft/ai-agents-for-beginners), used under the MIT License, Copyright (c) Microsoft Corporation. Study guides and quiz questions are written by Vikosha. Certificates are issued by Vikosha and are not issued or endorsed by Microsoft. Licence: MIT License, Copyright (c) Microsoft Corporation. View the original curriculum