AI coding agents are changing how software teams plan, build, test, and maintain code. Instead of using artificial intelligence only for one-off prompts, developers and technical teams can now work with agentic systems that help break down tasks, navigate codebases, suggest implementation steps, generate tests, identify issues, and support review workflows.
This course is designed to help learners use AI coding agents effectively and responsibly in real development environments. You will learn how to define agent-ready tasks, provide useful context, evaluate generated code, and apply human-in-the-loop practices that keep quality, security, and accountability at the centre of AI-assisted development.
This course is part of the Back-End and Intelligent Systems Development Certificate and gives learners practical experience with the workflows, review habits, and adoption considerations needed to use coding agents with confidence.
Who should enrol?
This course is intended for professionals who want to use AI coding agents in software, data, automation, technical product, or digital transformation work. It is especially relevant for:
- Software developers and engineers who want to improve productivity while maintaining code quality and accountability
- Technical leads and engineering managers responsible for team workflows, standards, AI adoption, or code review practices
- Data, automation, and IT professionals who create scripts, integrations, internal tools, or technical documentation
- Product managers, business analysts, and innovation professionals who work closely with technical teams and need to understand what coding agents can and cannot do
- Learners in software, AI, or digital transformation programs who want applied experience with agentic development workflows
What you will learn
- How coding agents differ from standard AI coding assistants and chat-based code tools
- How to define clear development tasks, constraints, context, and acceptance criteria for agentic coding workflows
- How to use coding agents to support planning, code generation, debugging, refactoring, testing, documentation, and review
- How to evaluate AI-generated code for correctness, maintainability, security, performance, and fit with project requirements
- How to design human-in-the-loop checkpoints that reduce risk and improve confidence in AI-assisted work
- How to create a practical adoption plan for using coding agents in an individual, team, or organizational setting
Skills you'll gain
By the end of this course, you will be able to:
- Set up an appropriate coding agent toolset for a development workflow
- Recognize and mitigate key risks when using coding agents
- Use coding agents effectively in editor-based, editor-less, and multi-worktree scenarios
- Manage agent swarms and parallel development workflows
- Design continuous delivery workflows involving coding agents
Course overview
This course uses applied examples and guided activities to help learners move from basic prompting to practical coding-agent workflows. Learners will explore how to plan a task, prepare context, work iteratively with an agent, review generated outputs, and decide when human judgment is required.
Activities may include workflow mapping, prompt and instruction design, code review exercises, test-generation activities, risk analysis, documentation review, and planning for responsible adoption.
Participants will also explore and compare popular coding agent tools and workflows, including Claude Code, OpenCode, Antigravity, and Codex.
Module outline
|
Module 1: Setting up a safe, productive environment |
|
|---|---|
|
Module 2: Generating quality code |
|
|
Module 3: The new workflow |
|
Recommended background
- Basic understanding of programming or software development workflows
- Experience with Python, JavaScript, TypeScript, web development, automation, or technical product work is helpful but the course is designed to focus on practical agentic workflows rather than advanced programming theory.
Why this course matters
Coding agents can help teams work faster, but speed is only valuable when quality, security, and accountability are protected. Organizations need professionals who can use AI coding tools thoughtfully, ask better questions, review outputs critically, and design workflows that support responsible adoption.
This course helps learners move beyond experimentation and into practical application. Participants will leave with stronger language, repeatable workflows, and a clearer understanding of how to use coding agents as part of a modern intelligent software systems development process
Other courses in the program
Frequently asked questions
What is a coding agent?
A coding agent is an AI-enabled development tool or workflow that can support software development tasks with greater autonomy than a standard coding assistant. Coding agents can help developers plan work, generate code, debug issues, review changes, and support testing workflows.
What is vibe coding?
Vibe coding refers to using AI-generated code rapidly and intuitively, often with minimal manual implementation. This course explores how to use vibe coding responsibly while understanding quality, security, maintainability, and governance considerations.
Do I need programming experience?
Yes. Learners should have a basic understanding of programming and software development workflows. Familiarity with Git, Python, JavaScript, or TypeScript is recommended.
Is this course self-paced?
Yes. This is a self-paced online course designed to support flexible learning schedules.
Which coding agents are explored in the course?
The course compares and discusses tools and workflows including Claude Code, OpenCode, Antigravity, and Codex.
Does the course cover responsible AI use?
Yes. Designed to focus on the practical skills needed to use coding agents as reliable development partners rather than unchecked automation tools, the course emphasizes human-in-the-loop practices, code review, risk mitigation, security, privacy, and responsible adoption strategies.