BrowserStack Meetup Dubai: Building Local AI Agents for Web Test Automation

About this event

Traditional web test automation often struggles with dynamic user interfaces and selector maintenance, consuming valuable QA time. Relying on black-box AI solutions can also feel opaque and costly, leaving teams without true control over their testing infrastructure. ​This hands-on workshop empowers QA engineers and developers to master self-hosted AI agents for web test automation. You'll build a practical, self-healing Playwright test suite from scratch using open-weights models. Gain code-level understanding and reduce your reliance on external services. ​Reserve your spot today to master the future of testing and reclaim control over your automation strategy. ​What Will You Learn • ​You'll build a working AI agent from scratch, understanding its core components: model, tool, loop, and stop condition. • ​You'll learn to get reliable, schema-checked output from self-hosted open-weights models, never relying on paid hosted services. • ​You'll master generating stable, verified Playwright locators from a live page and implementing self-healing test suites. • ​You'll grasp the real costs and benefits of running AI-assisted testing on your own infrastructure. ​Agenda • ​11:00 AM GST: Introduction & framing • ​11:15 AM GST: Setting up the harness • ​11:35 AM GST: An agent from scratch • ​11:55 AM GST: Moving to LangChain & LangGraph • ​12:15 PM GST: Break • ​12:35 PM GST: Agent 1 — the explorer • ​12:55 PM GST: Agent 2 — locators & test cases • ​01:20 PM GST: Agent 3 — the test builder • ​01:40 PM GST: The self-healing loop • ​01:55 PM GST: The full run & wrap-up ​What partcipants need to bring • ​A laptop with Python 3.11+, Node.js 20+ and the uv package manager installed. • ​The workshop repository, cloned and set up before the session. • ​A free Lightpanda Cloud API token, from console.lightpanda.io (https://console.lightpanda.io/?utm_source=luma). The token is shown only once, so generate and save it in advance. • ​Access to a model server (Ollama or llama.cpp): its address, a model name and an API key. (Optional - If you don't have one ready, we'll help you set it up on the day. ​Who Should Attend • ​QA engineers • ​SDETs • ​Developers who write automated tests • ​Python developers keen on AI in testing If you're ready to transform your approach to test automation and understand AI agents at a code level, this workshop is for you. Reserve your spot today.