AI-driven Software Engineering establishes a new paradigm in the software development lifecycle (SDLC), shifting from manual code creation to an iterative workflow of planning, AI-generation, and verification. It deals with the integration of Large Language Models (LLMs) into every stage of production, from architecture to deployment. Topics of the lecture include:
- Foundations of AI Development: LLM architecture for developers, effective prompt engineering, and the transition from 0-1 coding to AI-assisted iteration.
- Coding Agents: Architecture of autonomous coding agents, tool use, function calling, and the Model Context Protocol (MCP).
- The AI IDE: Context management in complex codebases, "Spec-driven" development, and integration of AI into Integrated Development Environments.
- Collaboration Patterns: Managing agent autonomy, human-agent interaction models, and "Human-in-the-loop" workflows.
- Modern Terminal & Automation: AI-enhanced Command Line Interfaces (CLI) and scripting automation.
- Automated UI Generation: GUI generation from design draft or natural language description.
- AI Security & Testing: Secure "vibe coding," SAST vs. DAST with AI, automated test suite generation, and detecting vulnerabilities in AI-generated code.
- Modern Support & Review: AI-assisted debugging, automated code reviews, and intelligent documentation generation.
- Post-Deployment: AI in Site Reliability Engineering (SRE), observability, and automated incident response.
- Dozent: Chunyang Chen
- Dozent: Ludwig Felder
- Dozent: Yuetian Mao
- Dozent: Christopher Zerbe