[100% Off] Agentic Ai Engineer Bootcamp
Build Python AI agents with tools, memory, RAG, MCP, multi-agent workflows, evaluation, and security.
What you’ll learn
- Build Python AI agents that use tools
- plan tasks
- track progress
- and stop within defined execution limits.,Add persistent memory and document retrieval to create assistants that answer questions with supporting sources.,Connect agents to MCP tools and coordinate specialist agents through workflows
- checkpoints
- retries
- and human approval.,Evaluate
- trace
- and secure an end-to-end research assistant with tests
- execution budgets
- and tool permission controls.
Requirements
- Basic Python knowledge
- including functions
- dictionaries
- and running scripts. You’ll need a computer with Python 3.10 or later
- a code editor
- and internet access for setup. No prior AI agent experience is required. Core exercises run locally; optional live model examples require an API key and may incur usage charges.
Description
This course contains the use of artificial intelligence.
The Agentic AI Engineer Bootcamp teaches you how to build AI agents that use tools, retrieve information, manage memory, and complete tasks through controlled workflows. Across 13 practical sections, you’ll progress from agent fundamentals to building and evaluating an end-to-end research assistant.
You’ll begin by understanding how AI agents differ from fixed workflows and how to choose a useful first project. Then, you’ll set up your Python environment, define response contracts, validate structured outputs, and troubleshoot model requests.
As you advance, you’ll learn to connect agents to tools, validate arguments before execution, handle failures, and make safe retry decisions. You’ll build execution loops with task planning, progress tracking, and stopping conditions. You’ll also explore conversation state, persistent memory, and how to select, update, retrieve, and delete remembered information.
The course introduces retrieval-augmented generation, including document chunking, source tracking, grounded answers, and handling missing or conflicting evidence. You’ll explore Model Context Protocol, specialist-agent handoffs, shared context, and workflows with checkpoints, recovery paths, and human approval.
You’ll also learn how to assess agent performance through evaluation datasets, practical rubrics, and regression checks. Observability lessons cover execution traces, token usage, latency, and execution budgets. Security lessons address prompt injection, untrusted content, secrets management, and tool permission boundaries.
Each section combines focused lectures with hands-on labs, quizzes, and coding exercises. Downloadable resources include lecture materials, worksheets, worked examples, starter code, and answer guides.
For your capstone, you’ll integrate these skills into a research assistant and test both successful outcomes and failure cases.
Basic Python knowledge is recommended. Core exercises run locally without paid model calls; optional live integrations require separate setup and may incur API usage charges.
