AI is making it easier than ever to ship features. Agentic coding tools can generate code, tests, APIs, UI flows, infrastructure changes, and pull requests at a speed that most engineering organizations have never experienced before. But this creates a new failure mode: teams may soon ship features faster than they can understand, operate, support, recover, and improve the products they are building.

Great product engineering has never been only about implementing the feature. It is also about designing how the product behaves in production: observability, business metrics, user journey tracking, feature analytics, supportability, failure reconciliation, auditability, rollback, release safety, incident response, and continuous learning. These are often treated as operational afterthoughts, but they are what determine whether a product can actually run with excellence.

This talk introduces the idea of AI Operable Products: products designed so both humans and AI systems can observe, reason about, and improve them in production. We will explore how product teams, engineers, AI engineers, SREs, and platform teams can move from “AI-assisted feature delivery” to a model where operability is designed into the product from day one.

Drawing from real-world experience leading Engineering, Digital Operations, SRE, and Cloud Infrastructure for digital banking systems serving more than 16 million customers, this talk will show why the next bottleneck is not code generation. The next bottleneck is operability.

Key takeaway: Agentic engineering helps teams build faster. AI Digital Operations helps teams operate better. The future belongs to teams that can do both.