Kseniia Korostelova

How do you deliver with AI and still sleep at night? By engineering trust into your delivery system.
In this talk, I'll share an end-to-end AI engineering workflow we built and ran on a production codebase, redefining roles, responsibilities, quality gates, and human decision points across the entire delivery lifecycle.
You'll leave with a practical framework for engineering AI workflows where reliability is designed in, trust is earned, and success becomes the default outcome, not a fortunate accident.

Everyone is looking for AI metrics. Organizations are tracking token consumption, AI adoption rates, generated code volume, and agent activity in an attempt to measure engineering performance. Yet, many teams still struggle to answer a simple question: what does quality look like with AI?
This talk argues that AI changed how software is produced, not what good software looks like. We'll explore why many AI KPIs miss the point and how to distinguish productivity, quality, and business outcomes when measuring engineering success.

In this talk, Kseniia Korostelova shares the results of a real experiment where she attempted to delegate parts of her engineering leadership work to AI. From architecture decisions and feature planning to design reviews and performance feedback, some tasks worked surprisingly well while others failed completely.
These experiences led to the concept of Organic Leadership: a leadership approach that uses AI to remove noise while preserving the human judgment, trust, and context that teams rely on.
You will learn where AI can genuinely augment engineering leadership today, where it cannot, and why the rise of AI may actually increase the value of truly human leadership.