--- title: "Who Tests AI-Written Code? Inside TesterArmy With Szymon Rybczak" description: "Szymon Rybczak explains how TesterArmy tests web and mobile apps with AI agents, creates test plans for pull requests, and fits into QA workflows." url: https://www.callstack.com/podcasts/who-tests-ai-written-code-inside-testerarmy-with-szymon-rybczak updated: 2026-10-24T08:39:44.301Z --- # Who Tests AI-Written Code? Inside TesterArmy With Szymon Rybczak Szymon Rybczak explains how TesterArmy tests web and mobile apps with AI agents, creates test plans for pull requests, and fits into QA workflows. _Guest: Szymon Rybczak, Co-founder & CEO @ TesterArmy · Host: Aleksandra Desmurs-Linczewska, Software Engineer @ Callstack · Published 2026-10-12 · Duration 12:52 · Category: Testing_ Listen: [YouTube](https://www.youtube.com/watch?v=Y-NTZvu6rkM) Szymon Rybczak, Co-founder & CEO of TesterArmy, joins us in the studio at Agent Conf 2026 in Warsaw to discuss AI testing for web and mobile apps and his journey from Callstack to Y Combinator. We talk about why manual QA struggled to keep pace with agent-written code, how TesterArmy creates test plans around pull requests, and the choices behind its managed platform and open-source tooling. Szymon also shares what it was like to build a company while finishing high school and why his first hire was a designer. #### Building a company while finishing high school Szymon describes founding TesterArmy during his final year of high school, studying for exams with friends on Discord while attending Y Combinator. His co-founder, Oskar Kwaśniewski, also previously worked at Callstack. Their experience building open-source libraries and developer tools shaped the company. After trying consumer apps, they returned to an area they knew from their own work: the gap between shipping code and checking that it works. #### AI startups, hardware, and robotics at YC We ask whether everyone at Y Combinator is building with AI. Szymon shares his impressions of the batch and describes the hardware and robotics startups he encountered. He discusses why investors are watching robotics for its own breakthrough moment. The conversation offers his perspective as a founder going through the accelerator alongside teams working on different parts of AI. #### Testing web and mobile apps with agents Szymon explains how manual QA became a bottleneck as teams used agents to write more code. He and Oskar wanted a testing tool developers could try themselves and evaluate before bringing it to their managers. TesterArmy started with web testing because browsers were faster to set up and iterate on. Szymon describes its expansion to native and React Native apps on iOS simulators and Android emulators, with the same goal: checking changes whether a person or an agent wrote the code. #### Test plans that follow the pull request Teams can start with a website URL or an uploaded app binary, then provide business-specific knowledge so the agent can distinguish an intended behavior from a bug. Szymon also explains how the platform uses a pull request's diff and description to create a test plan for that change. That gives teams a way to check a new feature alongside their existing tests and add those checks to their regular testing group. #### A managed platform and open-source testing tools We ask how much control users have over the underlying model. Szymon explains that TesterArmy manages model selection and infrastructure within its hosted product. He also outlines his plans for e2e, an open-source TypeScript SDK for building tests with a team's own model credentials and infrastructure. He says the SDK will use [agent-device](https://oss.callstack.com/agent-device/) under the hood for mobile testing. The conversation covers the different needs of teams that want a managed service and developers who want to build their own agents. #### A designer as the first hire and better tools for QA TesterArmy's first hire was a designer who created its pixel-art logo and animation. Szymon explains why brand and product quality deserved that investment, and how coding agents let the designer contribute to the codebase. We also ask what agentic testing means for QA roles, and Szymon describes working with QA teams and building tools they can use in their work. #### About the guest Szymon Rybczak is Co-founder & CEO of TesterArmy, a Y Combinator-backed company building AI testing tools for web and mobile apps. He previously worked at Callstack on React Native infrastructure and open source, including react-native-ai. #### Resources - [TesterArmy](https://tester.army/) - [e2e on GitHub](https://github.com/tester-army/e2e) - [Szymon Rybczak](https://szymonrybczak.dev/) - [Agent Conf](https://www.agent.sh/) Insights ## Learn more about Testing Here's everything we published recently on this topic. [View all](https://www.callstack.com/insights) - [How AI Agents Can Cut Manual QA by 30%](https://www.callstack.com/blog/how-ai-agents-can-cut-manual-qa-by-30) (Oct 30 · Article): Explore an agentic QA workflow targeting 30% less manual effort across mobile, web, and TV, with test selection, failure triage, and validation. - [Give QA Agents the Controls They Need With Appduct](https://www.callstack.com/blog/give-qa-agents-the-controls-they-need-with-appduct) (Oct 20 · Article): Use Appduct to give AI QA agents access to app functions for setting test conditions, triggering events, and checking results with UI automation. - [Agentic Mobile Testing With e2e and agent-device](https://www.callstack.com/blog/agentic-mobile-testing-with-e2e-and-agent-device) (Oct 13 · Article): See how e2e uses agent-device for AI-driven iOS and Android tests, combines exact assertions with agent actions, and replays cached steps without model calls. - [Automating Mobile QA With Cloud Agents](https://www.callstack.com/blog/automating-mobile-qa-with-cloud-agents) (Aug 4 · Article): Code review and green CI checks do not prove a mobile app still works. Learn how cloud agents, agent-device, and runtime evidence can verify app behavior before merge. - [Behavior Review for AI-Native Delivery](https://www.callstack.com/events/behavior-review-for-ai-native-delivery) (Jul 21 · Webinar): AI can increase code output faster than teams can verify the real app. See why behavior review matters before merge. - [Introducing Sniffler: Run only the E2E tests your changes can affect](https://www.callstack.com/blog/introducing-sniffler-run-only-the-e2e-tests-your-changes-can-affect) (Jul 6 · Article): Sniffler maps changed files to affected E2E tests, so teams can avoid running the full suite on every pull request. - [Mobile QA with Cloud Agents at AI Coding Summit London](https://www.callstack.com/events/mobile-qa-with-cloud-agents-at-ai-coding-summit-london) (Jul 6 · Talk): Michał Pierzchała will speak at AI Coding Summit London 2026 about mobile QA agents for iOS and Android pull request workflows. - [Automating Mobile QA with Cloud Agents](https://www.callstack.com/events/automating-mobile-qa-with-cloud-agents) (Jun 17 · Talk): Michał Pierzchała will present Automating Mobile QA with Cloud Agents at Ship London on 17 June 2026. - [How AI Is Changing React Native Development and Testing](https://www.callstack.com/podcasts/how-ai-is-changing-react-native-development-and-testing) (May 13 · Podcast): Zack Ebenfeld joins React Universe On Air to talk about how AI is changing React Native workflows, testing, QA, and day-to-day engineering work.