Stream: Virtual Room 7
Time: 11:15 - 12:00
The rise of AI coding assistants has introduced a new style of development often called “vibe coding” — describing a problem in natural language and letting AI generate the code. While this can dramatically accelerate development, it also raises important questions for teams responsible for mission-critical enterprise systems. AI tools are optimized to produce working code quickly, but not necessarily code that is secure, maintainable, or aligned with enterprise standards. Generated code can introduce hidden risks such as vulnerable dependencies, missing validation, or security weaknesses that may only surface later in production. This session explores where AI-assisted development fits within enterprise engineering practices. It examines the spectrum from manual coding to AI-assisted workflows, highlights common risks in AI-generated code, and discusses practical approaches for using these tools responsibly. We will also look at the broader skills shift already underway. As AI increasingly automates routine coding tasks, developer value is moving toward architecture, system understanding, security awareness, and the ability to guide and review AI-generated solutions effectively.
Vaidehi is an Associate Product Developer at BMC Software, where she works in AI/ML integration with mainframe and IMS environments. She completed her B.Tech in Computer Science in 2025 and joined BMC as a full-time developer after two internship stints with the company starting in 2024. Her work sits at the intersection of modern AI capabilities and enterprise mainframe systems - exploring how technologies like agentic AI and large language models can enhance traditional mainframe workflows. She is passionate about exploring emerging AI concepts while building a strong foundation in mainframe technologies.
Click here to give some Feedback so we can make it even better next year!