AI coding assistants can help streamline software development when used with clear instructions and appropriate oversight. This course introduces the principles behind language models, prompting, tools and agents before exploring GitHub Copilot in Visual Studio Code and its practical role within modern development workflows.
August sale - up to 15% off training courses – use code: EARLY0826TSG
View courses
Coding with AI
Develop confidence with AI-assisted coding
Build a more structured approach to AI-assisted coding. You'll learn how prompts and context influence code suggestions while developing the skills to review AI-generated output critically, helping you decide whether it meets your organisation's quality and security expectations.
Learning objectives
- Explain how language models contribute to AI-assisted software development
- Distinguish between prompts, tools, agents and supporting context
- Select an appropriate model for a defined coding activity
- Configure and operate GitHub Copilot within Visual Studio Code
- Shape coding requests by supplying clear instructions and useful context
- Inspect and refine code produced by an AI assistant
- Adapt Copilot through reusable prompts and custom agents
- Recognise where AI and automation require tighter control
- Assess threats involving exposed information, prompt injection and dependencies
- Apply safeguards that support secure AI-assisted coding
Key facts
Certification
This is a skills-based course with no formal accreditation.
Who it's for
The course is designed for developers, engineers and technical professionals using AI to support software development, as well as those responsible for teams producing code.
Prerequisites
You should be comfortable writing code without assistance. This isn’t essential when attending solely to understand the main concepts, potential risks and appropriate safeguards.
Exam information
There is no exam for this course.
Optional extras
There are no optional extras available for this course.
Pre-course
There is no pre-course work required.
Course syllabus
The Coding with AI syllabus covers the following topics.
- Language models and agents
- Models
- Prompts
- Prompt engineering
- LLM characteristics
- Thinking and reasoning
- Which model?
- Context
- Context window
- Tools
- What is an agent?
- The agent loop
- Agent types
- Subagents
- Memory
- GitHub Copilot in Visual Studio Code
- About GitHub Copilot
- Setup
- Suggestions
- Agents (chat)
- Prompting
- Context
- Tools
- Reviewing
- Editing previous chat requests
- Checkpoints
- Chat participants
- Inline chat
- Generate code context menu
- Slash commands
- Smart actions
- Switching accounts
- Signing out
- Disabling AI features
- Copilot customisation
- Reusable prompts
- Custom instructions
- Custom agents
- Model Context Protocol (MCP)
- Copilot security
- Execution and access
- Supply chain dependencies
- Approval automation
- Information exposure
- Prompt injection
- Trust boundaries
- Controlled scope
- Transparency
- Secrets management
- Recommended security baseline
- Enterprise policies
FAQs
This course explains how GitHub Copilot can support software development in Visual Studio Code, covering prompting, customisation, code review and approaches for managing quality and security risks.
What is AI-assisted coding and how does it work?
AI-assisted coding uses tools such as GitHub Copilot to support software development by generating or refining code in response to your instructions. The tool draws on the prompt and any available context to produce a suggested output. Human review remains important, as the resulting code must still be checked for quality and security.
Will I learn how to write better prompts for coding?
Yes. The course examines how the wording and structure of a prompt can affect the response produced by an AI coding tool. You’ll practice defining the task precisely and providing the context and constraints needed to produce more useful coding suggestions, helping you create requests that are better aligned with the coding task and development environment.
What level of coding experience do I need?
You should be comfortable writing code without assistance. This is not essential when attending solely to understand the main concepts and potential risks. This means the course is suitable both for professionals using AI while coding and those overseeing development teams.
How does context affect AI-generated code?
Context gives an AI coding tool more information about the task, codebase or intended outcome. Relevant context can help the tool produce suggestions that are better aligned with the task you're trying to complete. Missing or unclear information may lead to less useful outputs, which is why the course covers both effective input and careful review.
Why study with TSG?
25+ years' experience
Trusted by individuals, businesses and public sector organisations for over 25 years.
20,000+ learners trained
Join a network of learners focused on developing their software testing skills
98% customer satisfaction
TSG trainers are frequently rated as ‘good’ or ‘excellent’ by our learners