AI Coding Agents: From Vibe Coding to Agentic Engineering

Master Cursor, Claude Code, MCP, Agent Teams,GSD & AI coding workflows to build real-world software with modern AI agent

New

Created byYanni Goyal

Last updated 8/2026

English

What you'll learn

Understand AI coding agents, context engineering, agentic workflows, and modern AI-assisted software development.
Build real-world software using Cursor, GitHub Copilot, Codex, Antigravity, Claude Code, and other AI coding tools.
Master Claude Code with commands, sessions, Git workflows, MCP, Skills, Plugins, Hooks, Sub-Agents, and Agent Teams.
Build full-stack applications with AI agents using Next.js, FastAPI, Docker, SQLite, APIs, authentication, and AI models.
Use GitHub, MCP, Agent Teams, GSD, and spec-driven development to plan, build, test, and manage complex software projects.
Apply professional AI engineering practices to debug, review, document, secure, and improve AI-generated code.
Run AI coding agents locally, remotely, and in cloud sandboxes while understanding safe and autonomous development workflows.
Develop the skills and mindset needed to move from vibe coding to professional agentic engineering.

Course content

17 sections • 84 lectures • 12h 16m total length

Build a Zombie Survival FPS Game with Cursor AI13:00
Understanding the New Era of Software Development7:16
Course Roadmap: From AI Coding to Agentic Engineering5:08
Vibe Coding, Agentic Coding & Modern AI Development10:15
The 8 Levels of AI Coding Maturity9:37
How AI Coding Agents Actually Work: LLMs, Memory, Tools & Loops9:53
Context Engineering: The Most Important Skill in AI Coding11:34
Mastering agents.md: Building Better AI Coding Agents11:03
AI Coding Workflows: From YOLO Mode to Agent Teams15:04
Choosing the Right LLM: Beyond the AI Hype11:27

Hands-On with Cursor: Setting Up Your First AI Coding Project7:54
Exploring AGENTS.md & Cursor Settings for Smarter AI Coding9:56
Building SprintBoard with Cursor AI: Planning, YOLO Mode & Iteration13:14
Building SprintBoard with GitHub Copilot: Planning, Debugging & Best Practices11:43
Building SprintBoard with OpenAI Codex: Can It Beat Cursor & Copilot?11:47
Building SprintBoard with Google's Antigravity IDE & Gemini10:16
Cursor vs Copilot vs Codex vs Antigravity: Which AI Coding Tool Wins?4:10
How to Choose the Right LLM for AI Coding Agents7:06

The AI Engineer Mindset: 5 Rules That Save Hours of Debugging6:33
Building AI Projects the Right Way: OpenRouter & Best Practices5:31
YOLO Mode: Building a Professional Portfolio Website with Cursor AI11:04
Building an AI Digital Twin & Publishing Your Project to GitHub7:16
Learn Faster with AI: Generate Documentation, Review Code & Improve with Multipl6:53
AI Coding Reality Check: Why Good Engineers Get Better Results from AI10:35

Web Apps Explained: Frontend, Backend, APIs & Docker11:23
Setting Up TaskPilot: From Frontend Demo to Full-Stack AI App12:31
Transform the TaskPilot frontend demo into the foundation of a full-stack AI application. Review the existing project, understand the target architecture, and prepare the codebase for backend APIs, persistence, authentication, Docker, and AI integration.
Planning TaskPilot with an AI Coding Agent: From Requirements to Architecture13:50
Building TaskPilot Step-by-Step: Docker, FastAPI & Authentication11:12
Building TaskPilot Step-by-Step: Database, APIs & Persistent Frontend16:05
Completing TaskPilot: OpenRouter, AI Assistant & Course Progress18:47

From Vibe Coding to Vibe Engineering12:17
The Rise of Claude Code: History, Setup & Installation in VS Code9:52
Getting Started with Claude Code CLI: Init, Context & Testing10:56
Claude Code Review: Fixing Hallucinations & Refactoring Code9:53
Claude Code Beyond Anthropic: OpenRouter & Local AI Models5:51
OpenCode Deep Dive: Free AI Models, Multiple Providers & Real-World Coding4:43
Mastering Claude Code: Essential Commands, Shortcuts & Productivity Tips7:51
Mastering Claude Code Sessions, Checkpoints & Git Workflows8:00

Building the First PactForge Feature with FeatureDev10:15
Debugging AI-Generated Code: A Disciplined Workflow5:58
Mastering CLAUDE.md: Teaching Claude Your Coding Style6:24
Writing CLAUDE.md and Creating Custom Skills6:26
Planning the V1 Roadmap with GitHub Issues6:24
Building the V1 Foundation with Claude Code & FeatureDev6:55
Testing the AI Legal Document Generator Built with Claude Code8:44
Completing PactForge: Authentication, Final Merge & Project Wrap-Up16:33

Requirements

Basic programming knowledge is recommended. No prior experience with AI coding agents, Claude Code, Cursor, MCP, or agentic development is required.
A computer capable of running modern development tools and a willingness to experiment with AI-assisted software development.

Description

AI coding is evolving rapidly—from simple AI-assisted programming and vibe coding to autonomous agents, multi-agent teams, and spec-driven software development. This course takes you through that entire journey.

You’ll learn how modern AI coding agents actually work, how to engineer context effectively, and how to use AI to plan, build, debug, test, review, and improve real-world software projects.

Starting with the foundations, you’ll explore vibe coding, agentic coding, AI coding workflows, context engineering, LLM selection, and the mindset needed to work effectively with AI. You’ll then get hands-on with modern tools including Cursor, GitHub Copilot, OpenAI Codex, Google Antigravity, Claude Code, and OpenCode.

The course goes deep into Claude Code, covering commands, sessions, Git workflows, MCP servers, Skills, Plugins, Hooks, Sub-Agents, Agent Teams, sandboxing, remote development, and the Claude Agent SDK. You’ll also learn how to connect AI agents with GitHub and build more autonomous development workflows.

You won’t just watch demonstrations—you’ll build real projects. You’ll create full-stack applications with technologies such as Next.js, FastAPI, Docker, SQLite, APIs, authentication, and AI models, while applying AI engineering practices throughout the development process.

You’ll also explore multi-agent development, specialized AI agents, GSD, spec-driven development, and agent orchestration. By the end, you’ll understand how to move beyond basic vibe coding and use AI coding agents as powerful engineering partners.

Whether you’re a developer looking to work faster with AI or an engineer preparing for the next generation of software development, this course gives you the practical knowledge and hands-on experience to move from vibe coding to agentic engineering.

Who this course is for:

Developers who want to learn modern AI-assisted software development and move beyond traditional coding and basic vibe coding.
Software engineers who want hands-on experience with Cursor, Claude Code, Codex, Copilot, MCP, Skills, Agent Teams, and GSD.
Developers interested in building real-world full-stack applications with AI coding agents and modern agentic engineering workflows.
Programmers who want to understand how AI coding agents can be used for planning, coding, debugging, testing, GitHub workflows, and autonomous development.
Learners who want to progress from AI coding fundamentals to advanced multi-agent and spec-driven development.

AI Coding Agents: From Vibe Coding to Agentic Engineering

What you'll learn

Explore related topics

Course content

AI Coding Foundations10 lectures • 1hr 44min

Hands-On with Modern AI Coding Tools8 lectures • 1hr 16min

The AI Engineer Mindset & Best Practices6 lectures • 48min

Building TaskPilot: From Demo to Full-Stack App6 lectures • 1hr 24min

Mastering Claude Code8 lectures • 1hr 9min

Autonomous AI Coding2 lectures • 16min

Extending Claude Code with MCP, Skills & Plugins3 lectures • 22min

Advanced Claude Code: Skills, MCP & Plugins4 lectures • 28min

AI-Powered Software Development with GitHub4 lectures • 28min

Building PactForge with AI Coding Agents8 lectures • 1hr 8min

Requirements

Description

Who this course is for: