Participants learn how to apply practical AI techniques across the entire software development lifecycle – from code generation and testing to debugging and documentation, in their selected application area and programming language of choice.
Course Topics:
- Introduction to AI-Assisted Development: AI Coding Assistants Overview (GitHub Copilot, Claude Code, ChatGPT, Cursor), understanding LLM capabilities and limitations, architecture (context windows, prompt engineering), and establishing baseline productivity measurements for ROI tracking.
- Code Generation and Completion: Effective prompt engineering, context management, and domain-specific generation (mentored sessions for the specific end context depending on the participant background, e.g. embedded drivers, protocols, web REST APIs). Adapting AI-generated code to match strict project conventions and standards.
- Automated Testing with AI: Accelerating Test-Driven Development (TDD) using AI-generated test cases and assertions. Creating platform-specific tests to achieve 90%+ coverage.
- Code Review and Refactoring: Using AI as a first-pass reviewer for pattern detection and identifying code smells. Legacy code modernization strategies across platforms.
- Documentation Generation: Automating inline documentation (Doxygen, JSDoc or similar). Generating architecture documentation (READMEs, OpenAPI/Swagger specs) and diagrams (C4 model, PlantUML) directly from code.
- Full Implementation Exercises: Building complete modules from requirements depending on the participants domain expertise, tracking time savings, and documenting ROI.
- Capstone Project: In a specific area of expertise (through groupwork) participants use AI to create a software module (e.g. embedded code – modernizing a legacy codebase; web project – building a feature-complete module; evaluated on code quality, test coverage, documentation completeness, and productivity metrics.
Requirements
Software: Chrome browser, and access to AI coding assistants (e.g., GitHub Copilot, Claude, ChatGPT).
Hardware: PC with an internet connection.
Prior knowledge: Participants should have a basic engineering background, preferably a software background, and knowledge of at least one programming language.
