Booklet

Qase Perspectives on AI in Testing

Two arguments about quality in the agentic era.

Download the booklet (PDF)

From the diagnosis to the machinery

Qase Perspectives on AI in Testing is the series where we publish what we are learning about software quality under AI-accelerated development. This edition collects the first two papers in one volume. Each stands on its own. Read in order, they run from the diagnosis to the machinery.

The two parts take different approaches to the same problem. Part One is a research argument: what erodes when agents do the building is human understanding, and no amount of inspection restores it after the fact. Part Two is a product argument: verification is the bottleneck, and it needs a shared layer that every agent works through.

What's inside

Part One

Nobody carries the code

Why testing-after breaks under AI-accelerated development. A research paper.

The diagnosis. What erodes under AI-accelerated development is human understanding, and no amount of inspection restores it after the fact. Comprehension debt, intent debt, the generative ratification loop, and one testable proposal for preventing all three.

Vitaly Sharovatov, Developer Advocate at Qase, with Liliia Abdulina, Head of QA, JetBrains

Read Part One on the blog

Part Two

Test management for AI coding agents

The shared quality layer: the context and controls agents use to verify their work inside your project.

The machinery. Generation got cheap and verification did not, so verification is the bottleneck, and it needs a shared layer that every agent works through: one project context, QA-owned rules, and a governed way to check every change.

Maksim Koutun, VP of Engineering, Qase

Read Part Two on the blog

Written for QA leads, engineering managers, and anyone deciding how their team works alongside coding agents. Print it, share it, or read it on screen.

Your browser can't display the PDF preview. Open the booklet

Ship quality software faster with Qase