AI Engineer – Computer‑Use Agents

About the role

We are building an enterprise AI platform where agents operate business applications through their user interface, just like a person would. You will teach agents to navigate real‑world web portals, keep them functional as sites evolve, and ensure business users can rely on the outcomes.

Key responsibilities

• Teach agents to navigate new web portals, handling navigation, search, filtering, pagination, and document download.
• Manage complex site interactions such as multi‑factor authentication, broken searches, inconsistent categories, and unexpected pop‑ups.
• Monitor agent runs, diagnose failures from traces and replays, and implement fixes.
• Detect portal changes and collaborate with the Agent Learning & Evaluation engineer to adapt agents without breaking.
• Build automated tests against recorded sessions to validate changes before production.
• Work closely with business users to understand success criteria and translate feedback into improvements.

Required profile

• 6+ years of software engineering experience, including at least 1 year building LLM agents or large‑scale browser automation.
• Strong debugging skills and ability to troubleshoot issues on sites you do not control.
• Experience with Arabic‑language websites is a plus.

Required skills

• Browser automation frameworks: Playwright, Puppeteer, Selenium.
• Deep understanding of the DOM, Chrome DevTools Protocol, network traffic, and authentication flows.
• Proficiency in Python and/or TypeScript.
• Knowledge of agent‑oriented browser frameworks and tool design.
• Experience with large‑scale web automation reliability (retries, rate limits, change detection).
• Familiarity with vision‑language models for UI agents (e.g., Qwen‑VL, UI‑TARS).