GuideBeginner
How Coding Agents Work
Understand the conceptual and practical fundamentals of how coding agents work: from LLM mechanics to the workflow that separates productive developers from the ones who lose time with AI. This isn't a guide about building production agents — that's AI Engineering — it's about understanding how they think, act, and fail, so you can direct them with professional judgment using any tool: Claude Code, Cursor, Copilot, Cline, or whatever comes next. The seven modules move from the paradigm shift in software development, to LLMs explained for developers (not researchers), to the agentic loop and tool calling that turn a model into an agent, to the toolbox an agent uses to interact with your code, to the mental models for directing human-agent collaboration, and to the Research → Plan → Execute → Validate workflow. It closes with a hands-on project: building a mini coding agent from scratch to understand the agentic loop from the inside.
- 33
- lessons
- 7
- modules
- English · Spanish
- available in
- Yes
- certificate
- Free
- access
Outcomes
What you'll be able to do
- Explain what changed in software development with the arrival of coding agents, and tell apart vibe coding, traditional development, and agentic development
- Understand LLMs from a developer's perspective: tokens and context windows, next-token prediction, why they hallucinate, and what they can and can't do
- Precisely define what a coding agent is (LLM + tools + loop + autonomy), and trace the agentic loop — observe → think → act → observe — with tool calling and the ReAct pattern
- Tell a chatbot apart from a real coding agent with technical judgment, beyond the marketing
- Name and use an agent's tool categories (files, shell, web), understand how it explores your codebase, and configure context and permissions with files like CLAUDE.md, AGENTS.md, or .cursorrules
- Apply human-agent collaboration mental models ("power tool", "intern manager"), act as a circuit breaker, and calibrate your trust in the output based on the task's risk
- Critically analyze the METR study on AI productivity — with its data and its limits — instead of settling for the headline
- Apply the Research → Plan → Execute → Validate workflow and its variants, and recognize the most common anti-patterns: vibe coding, prompt-and-pray, context dumping
- Build a functional mini coding agent in Python with real tool calling and logging of every decision, and document the insights you observed about how it reasons
Before you start
What you need to bring
It's for you if...
- Developers who use a coding agent daily without understanding what happens under the hood when they give it a prompt
- Developers who haven't adopted any tool yet and want the fundamentals before choosing one
- Senior devs who need to evaluate coding agents with data — METR, Stack Overflow, DORA — instead of marketing hype
- Anyone who uses or is about to use Claude Code, Cursor, Copilot, or any coding agent, regardless of which one
Requirements and materials
- Basic programming knowledge in any language, and familiarity with an editor, terminal, and git
- No prior experience with any specific coding agent or Machine Learning knowledge required
- For the final project in Module 7: an Anthropic or OpenAI API key and basic Python (estimated total cost of a few dollars)
Content
The syllabus, module by module
Open any of them to see its lessons.
- Module 4: The Agent's Toolbox — How They Interact with Your Code
- File Operations, Shell, and Web Search: The Three Categories of Tools
- How Agents "See" Your Codebase
- Context Management: Large Projects and the Context Window Limit
- Configuration and Permissions: CLAUDE.md, .cursorrules, AGENTS.md, and Sandboxing
- Module 6: The Fundamental Workflow — Research → Plan → Execute → Validate
- Research → Plan → Execute → Validate: The Fundamental Workflow
- Workflow Variants: PRD → Plan → Todo → Code, Spec-First, Explore → Plan → Code
- Verification: The Step Everyone Skips
- Common Anti-Patterns: What NOT to Do with Coding Agents
Common questions
What people usually ask
No limit. It's a free guide: come in whenever you like, as often as you like.
No. Modules run from easier to harder, but you can jump to the one you need. Progress is saved per lesson.
Whatever is needed is listed under “What you need to bring”, above. If nothing is listed there, you can start from zero.
In the Club's WhatsApp group, and every two weeks there's a live with an instructor where questions get worked through.
Yes. It's issued automatically once you finish every lesson, with a verifiable code you can share on LinkedIn.
Start whenever you like
What students say
These reviews are from enrolled students who completed at least 50% of the course. We moderate reviews only on content grounds (spam, offensive language, personal data), never for being critical or negative.
No approved reviews yet.
Be the first to share your experience!