The common belief that improving code quality slows down development is a misconception that can be disproven with empirical data.
Studying psychology provides techniques to visualize software issues and communicate the impact of technical debt to non-technical stakeholders.
Techniques from criminal profiling, like geographical offender profiling, can be adapted to track developer behavior and identify critical code hotspots.
Version control data reveals organizational patterns, such as poor team-architecture alignment or knowledge silos like the 'lone wolf' developer.
The conflict between speed and quality arises from misaligned feedback loops, where new features offer immediate value while the costs of poor quality are delayed.
The rapid code generation enabled by AI tools increases the volume of code so quickly that managing technical debt becomes essential for organizational survival.
Instead of just generating new code, AI's greatest potential lies in helping developers analyze, understand, and refactor complex legacy codebases.
Developers can gain management buy-in for quality initiatives by presenting data-backed evidence and framing the problem in business terms like efficiency and time-to-market.
Poor code quality can waste up to 40% of engineering capacity, making it a critical metric that should be tracked as a KPI for the entire organization.
Since code is read far more often than it is written, the primary goal should be to create clear, understandable code for other developers, not just the machine.
31:39Rainer Stropek
59:16Ryan J Salva
47:35Laurie Voss, Rey Bango, Hannah Foxwell, Rizel Scarlett & Thomas Steiner
1:08:42Chris Heilmann & Daniel Cranney
44:20Cassidy Williams
1:08:42
04:17Robert Werner
15:08YK Sugi
Jobs that call for the skills explored in this talk.
autoiXpert GmbH & Co. KG
Stuttgart, Germany
Saby Company Page
Sarajevo, Bosnia and Herzegovina
Redactie Descherpenzeelgids.nl
Veenendaal, Netherlands