Summary
Disclaimer: This summary has been generated by AI. It is experimental, and feedback is welcomed. Please reach out to info@qconlondon.com with any comments or concerns.
Joy Ebertz discusses the increasing prominence of technical debt in the era of AI and its management. The focus is on identifying how to prioritize tech debt and creating effective business cases for addressing it. Additionally, the talk explores AI's role in problem magnification and resolution.
- Introduction:
- Background on Joy's professional and personal interests.
- Importance of managing technical debt effectively in AI-driven environments.
- Main Points:
- Six Questions for Prioritizing Tech Debt: These address potential costs if tech debt is ignored, opportunity costs, training and maintenance efforts, and broader impacts on users and developers.
- Building a Business Case: Transforming technical concerns into financial terms to compare with product features and justify tech debt resolution efforts.
- Role of AI:
- AI speeds up code production but also introduces lower-quality code (slop).
- AI tools ease understanding but may exacerbate existing technical issues rather than alleviate them entirely.
- Conclusion:
- Final thoughts on AI and tech debt, and a call to avoid perfectionism that may threaten business viability.
Q&A Session: Consideration of political and organizational challenges when addressing tech debt, highlighting the necessity of presenting a persuasive business case to stakeholders.
Reference: This summary is based on the presented concepts from Joy Ebertz's session on managing technical debt and leveraging AI responsibly.
This is the end of the AI-generated content.
Abstract
As we move into the era of AI, code that could be called slop or tech debt is increasing faster than ever. This means that managing that tech debt is just as important as ever. However, how do you prioritize which tech debt to work on? How do you convince others that any of it is important? In the world of AI, is it important? Our goal at the end, is not to eliminate tech debt, but to write the best software we can for the situation in which we find ourselves. In some cases this means more tech debt, in others, less. In this talk, I'll discuss how to think about prioritizing tech debt, how to create a business case for your tech debt, and how AI may or may not color any of this.
Speaker
Joy Ebertz
Principal Engineer @Imprint, Blogger, and Speaker, Previously @Harness, @Split, & @Box
Joy is a Principal Software Engineer at Imprint where she is currently trying to detangle their configuration. She is a primarily backend focused developer with extensive experience in configuration, microservices architecture, revamping authorization frameworks, creating REST API standards, audit logging and more. Prior to Imprint, she has worked at Harness, Split, and Box in Staff+ roles. In addition to designing software and writing a lot of code, she also maintains a blog: joy.ebertz.run/blogs.html (with both career and technical topics). In her free time, she does a lot of traveling, reading, and running ridiculously long distances (mostly on trails).
