Free demoJob-oriented courses at Debugshala, Old Palasia, Indore. Freshers, non-IT and year-gap welcome.Book a free demo +91-8982385539
Join the Course
Free course · by Debugshala, Indore

AI Agents Tutorial

19 free lessons, in the order we teach them in class: start at lesson 1 and work down. Every lesson has an analogy, a diagram, code you can run or copy, a quick quiz and an FAQ.

Start lesson 1 Learn it in class: free demo
  • 19Lessons
  • 3Modules
  • ~3 hReading time
  • FreeAlways

01Agent Foundations

  1. AI Agents Explained: LLM, Tools, Memory and the Agent LoopAI agents explained: how an LLM, tools, memory and a loop work together, the ReAct pattern, planning and guardrails, with a toy agent you can run in Python.
  2. AI Agent vs LLM vs RAG: Key Differences with ExamplesAI agent vs LLM vs RAG: what each one does, how they differ on knowledge, actions and cost, and how to pick the right one, with runnable Python simulations.
  3. Building Blocks of AI Agents: Role, Tools, Memory, GuardrailsThe six building blocks of AI agents: role, task, tools, memory, guardrails and cooperation, how each shapes behaviour, and a Python agent spec you can lint.
  4. 5 Agentic AI Design Patterns Explained with Python ExamplesThe 5 agentic AI design patterns, reflection, tool use, ReAct, planning and multi-agent, explained with when to use each and runnable Python simulations.
  5. 5 Levels of Agentic AI Systems: From Responder to AutonomousThe 5 levels of agentic AI explained: basic responder, router, tool calling, multi-agent and autonomous, who controls the flow at each, and how to pick a level.
  6. Memory in AI Agents: Short-Term, Long-Term, Episodic and MoreMemory in AI agents explained: short-term context, semantic, episodic and procedural long-term memory, and how agents save, retrieve and forget, in Python.
  7. Agentic AI Glossary: 40 Must-Know Terms Explained SimplyAgentic AI glossary: 40 must-know terms, from agent loop and tool calling to MCP, A2A, guardrails and agentic RAG, each in plain English with examples.

02Building Agents

  1. Build a ReAct Agent from Scratch in Python (No Framework)Build a ReAct agent from scratch in Python: write the prompt, parse Thought and Action text, run tools safely, stop loops, then plug in a real LLM.
  2. Tools and Function Calling for AI Agents: A Practical GuideHow function calling works in AI agents: JSON Schema tool definitions, the call and result cycle, safe validation, and a real API loop, with Python examples.
  3. Multi-Agent System Patterns: 7 Ways AI Agents Work TogetherMulti-agent system patterns explained: sequential, parallel, loop, router, aggregator, hierarchical and network, with handoffs, trade-offs and Python demos.
  4. Agentic RAG Explained: Build a Self-Correcting RAG AgentAgentic RAG explained: how an agent routes questions, grades retrieved passages, rewrites weak queries and falls back to other sources, with Python code.
  5. Guardrails for AI Agents: Input, Tool and Output ChecksGuardrails for AI agents explained: input filters, tool allowlists, human approval, output checks and run budgets, with Python code and an Agents SDK example.
  6. How to Evaluate AI Agents: Metrics, Test Sets and CodeHow to evaluate AI agents: outcome, trajectory, efficiency and safety metrics, test sets, pass@k vs pass^k and LLM judges, with a runnable Python harness.
  7. Deploying AI Agents to Production: Architecture and CodeDeploying AI agents to production: API, queue and worker design, streaming, checkpoints, retries, idempotent tools, tracing and costs, with Python examples.
  8. A2A and AG-UI Explained: Agent Protocols with Python ExamplesA2A and AG-UI agent protocols explained: agent cards, tasks and states, streamed UI events, how they differ from MCP, and Python code for both protocols.

03Agent Projects

  1. Build a Multi-Agent Research Assistant in Python: Full ProjectBuild a multi-agent research assistant in Python: planner, parallel searchers, cited writer and fact-checker, simulated offline, then a real Agents SDK version.
  2. Build a Voice RAG Agent in Python: Speech In, Answers OutBuild a voice RAG agent in Python: voice activity detection, speech-to-text, retrieval, speech-friendly answers and text-to-speech, simulated, then for real.
  3. Build a Multi-Agent Travel Finder in Python with LangGraphBuild a multi-agent travel finder in Python: parse a trip request, search flights and hotels in parallel, fit a budget, write an itinerary, then use LangGraph.
  4. Build a Financial Analyst AI Agent in Python with MCP ToolsBuild a financial analyst AI agent in Python: parse stock questions, compute return, volatility and drawdown in code, serve them over MCP, check every report.
Practise on Debugshala TestOnline compiler, coding questions, aptitude and reasoning MCQs. Free at test.debugshala.com ↗