Architecting a Cross-Platform AI Agent for End-to-End Testing | Bay Area Meetup

The Test Tribe 7th SFO Meetup – Architecting a Cross-Platform AI Agent for End-to-End Testing
End-to-end testing with AI is moving far beyond simple LLM wrappers and scripted automation. As applications span Web, iOS, and Android, building intelligent systems that can reliably execute long, complex test flows requires a new architectural mindset.

At the 7th SFO Meetup, we’ll dive deep into what it really takes to design a cross-platform AI testing agent-one that can execute 100+ step end-to-end scenarios, adapt to unexpected UI states, and operate across platforms using a single, unified intent model.

This session is a technical deep dive into agentic architecture, semantic understanding, and resilient automation-designed for engineers who want to push past demos and build systems that work in production.

Event Details

• Date: 25th February 2026 (Wednesday)
• Time: 6:00 PM – 8:00 PM PST
• Venue: 1631 N First St, Suite 200, San Francisco

Session: Architecting a Cross-Platform AI Agent for End-to-End Testing

Speaker: Alex Rodionov

About the Session
Building an AI agent capable of executing complex, multi-platform end-to-end tests requires more than prompting an LLM.
In this session, Alex Rodionov explores how to architect a sophisticated agentic system using Alumnium, designed to translate high-level natural language instructions-such as “click log in”-into precise, platform-specific actions across Web, iOS, and Android.

You’ll learn how the agent:

• Uses accessibility trees to understand UI structure
• Applies semantic pruning to overcome context window limitations
• Implements autonomous feedback loops to self-correct when real-world UI behavior deviates from expectations

The result is a resilient testing agent that can adapt in real time-bridging the gap between human-readable test intent and reliable automated execution.

Key Takeaways

• How to use Alumnium to decouple high-level natural language intent from low-level platform implementation
• How a single test suite can run across Web, iOS, and Android
• A practical demonstration of integrating agentic CLI tools like Claude Code to execute tests written entirely in prose
• How agentic workflows bridge the gap between manual test cases and automated execution
• Architectural patterns for building resilient, self-correcting AI agents

About the Speaker
Alex Rodionov is a software engineer with 15+ years of experience in high-scale quality engineering and build infrastructure.
He is a technical lead on the Selenium project and the author of Ruby language support in Bazel. As the creator of Alumnium, Alex focuses on bridging developer productivity and resilient automation through semantic, agentic workflows.

His work sits at the intersection of testing, infrastructure, and AI-pushing the boundaries of what modern automation systems can achieve.

Why You Should Attend

• Learn how to design AI agents that work beyond demos
• Understand cross-platform automation at an architectural level
• See how natural language can drive real, reliable E2E testing
• Gain insights from a core contributor to Selenium
• Connect with San Francisco’s QA, automation, and engineering community
• Be part of The Test Tribe’s 7th SFO Meetup

About The Test Tribe
The Test Tribe is the world’s largest software testing community, empowering testers globally since 2018. With 700+ events and 150K+ members across 130+ countries, we bring testers together to learn, grow, and connect through meetups, conferences, cohorts, masterclasses, and more.

By RSVPing, you agree to our Terms and Conditions and Privacy Policy and consent to be contacted by The Test Tribe and our event partners.