Software Engineer – Evaluation Engineer (AI‑Assisted Testing)

About the role

This is a newly created position in Pakistan focused on building test harnesses and automated pipelines that validate both AI‑driven and traditional product features. You will ensure that APIs, workflows, data pipelines, and UI flows work correctly before they ship.

Key responsibilities

• Design, develop, and maintain test harnesses for existing products and new features.
• Create scripted test suites to catch regressions, correctness issues, and edge cases across releases.
• Evaluate AI‑assisted or AI‑driven components (agents, AI‑generated code, AI‑assisted workflows) using harness‑based approaches.
• Collaborate with backend engineers working on a Node.js/React stack to improve testability and observability.
• Analyze test results, identify failure patterns, and produce clear reports for engineering stakeholders.
• Continuously enhance testing tooling and workflows as the product evolves.
• Stay up‑to‑date with AI‑driven development practices that increasingly affect testing.

Required profile

• 6 months to 2 years of professional software engineering experience.
• Strong grasp of data structures, algorithms, and problem‑solving.
• Hands‑on experience with Node.js and/or React in a production codebase.
• Interest or hobby‑level exposure to AI‑driven development tools such as GitHub Copilot, Claude Code, or Cursor.
• Analytical mindset capable of interpreting test output and extracting meaningful signals.
• High eagerness to learn in a role without an established playbook in Pakistan.
• Experience in product‑focused companies is a plus, but services or outsourcing backgrounds are also acceptable.

Required skills

• Node.js
• React
• JavaScript
• Test automation and scripting
• AI development tools (GitHub Copilot, Claude Code, Cursor)