Schedule

We'll talk about the schedule, sessions and share information.

A special talk is planned for this slot. We’re keeping the details under wraps for now, but we’ll reveal everything very soon.

How do you measure the real impact of GenAI in testing at the scale of 1,000+ QA engineers and separate actual value from the “wow” effect? We’ll share the metric framework and results from real implementations: LLMs for generating test documentation, accelerating E2E automation, and AI agents running regression tests.

We will tell you what problems appeared, how we learned to reproduce the defect, how we stepped on a rake, and how it all ended (or did not end).

Let's build a simple RAG bot based on Heisenbug conference program.

The practice of building an agent-based system for mobile regression: interaction with the device, model selection, and a set of engineering techniques that reduce non-determinism and help validate the result without self-deception.

Testing the Smile Payment service.

Security check and search for biometric system vulnerabilities.

We will discuss how to implement AI in a company if you do not have large resources and free ML teams. I'll show you the free Roo Code AI assistant, as well as an unusual approach to writing feature evaluation agents, commit analysis, and requirements testing.

I will tell you how to build a simple and convenient tool for measuring the energy consumption of various electrical appliances and integrate it into your infrastructure.

How to transform chaotic "vibecoding" into a controlled engineering process using Spec-Driven Development, custom skills, and proper testing tools. Through a live project build (Web, Mobile, Backend), I will demonstrate how to get AI agents to deliver high-quality results on the first try, avoiding the trap of endless chat corrections.

Let's systematize the approach to testing in game development. I'll show you how test design techniques are used in gamedev and share patterns of approaches to key game mechanics.

Poorly secured test repositories are not just a future threat that will materialize sooner or later. This is a current, actively exploited security breach. And as the protection of the production code strengthens, this gap will only widen, making attacks on tests more attractive to intruders. The question is not whether there will be an attack, but when it will affect your company.

Let's reflect and discuss new tools together!

LLM-as-a-Judge in AI agent testing automation — let's discuss why this practice is vital, which models to choose, why LLM-as-a-Judge will deceive you with its simplicity, and how to start writing regression automation for your AI agents.

Errors occur at all stages of software development. The sooner we find out, the cheaper it is to fix them. At WS, testers will learn how to identify errors in requirements before the development stage. This will reduce the number of potential errors, the time and effort of the team to fix them.

The classic "how to reproduce this?" dilemma has been resolved! This talk introduces my browser extension that handles action recording, captures errors, and helps to produce bug reports automatically — no more manual note-taking or relying on memory.

This talk focuses on the practice of testing applications based on large language models (LLM) using the DeepEval tool. It will cover the automation of quality assessment using the LLM-as-a-Judge approach, specialized metrics, and the integration of testing into the development process to ensure the reliability and predictability of system operation.

An AI is losing context during codebase growth. I will demonstrate a few ways to manage it and make your work more comfortable.

A special talk is planned for this slot. We’re keeping the details under wraps for now, but we’ll reveal everything very soon.

Summing up the results of the conference, remembering the highlights and talking about plans.