3226: How Instabug Uses AI to Catch Bugs Before Users Do

What if your mobile app could detect bugs, fix UI inconsistencies, and spot user frustration before a user ever reports it? In today’s episode, I sit down with Kenny Johnston, Chief Product Officer at Instabug, to explore how AI is reshaping the way developers build, test, and maintain mobile apps.

Instabug is taking mobile observability to an entirely new level by developing what Kenny describes as “zero maintenance apps.” Powered by on-device AI models, their platform can now detect subtle UX breakdowns, visual design flaws, and even frustration signals that wouldn’t normally trigger crash reports. Whether it’s an unresponsive button, a layout shift, or a broken navigation path, Instabug flags the issue, often before a user ever notices.

Kenny shares how Instabug’s approach to AI is helping development teams move faster and smarter, particularly in high-stakes environments like retail and e-commerce where performance peaks during events like Black Friday or Valentine’s Day. Through real-time crash reporting, automated UI analysis, and deep session insights, developers can spot and solve problems that would otherwise get lost in a backlog or surface in app store reviews.

We also explore the unique pressures of mobile development. With no quick rollbacks and high user expectations, developers need tools tailored to the realities of app store approvals, device fragmentation, and version-specific bugs. Instabug’s platform brings together observability, feedback, and issue reproduction in a way that simplifies the mobile stack and accelerates release cycles.

Kenny draws on his experience at GitLab to reflect on the need to consolidate tools and workflows in mobile development. He offers valuable insights for product leaders and mobile engineers on how to navigate change, evolve their approach, and stay curious in the face of constant technical demands.

So how can your team shift from reactive debugging to proactive experience design? And are you really seeing all the issues your users encounter or just the ones they report? It's time to find out.