Senior Software Engineer, AI Training - Ireland

Role overview

This is a contract engagement focused on evaluating modern AI coding agents rather than writing production code. The work centers on assessing whether AI-generated interactions reflect strong engineering judgment, useful reasoning, and trustworthy explanations. It is a part-time, fully remote opportunity suited to senior engineers who can make rigorous, opinionated judgments about AI-assisted development workflows.

Responsibilities

- Evaluate end-to-end AI-generated coding interactions for usefulness, high-level correctness, and alignment with how experienced engineers think - Judge the quality of explanations, reasoning, and preamble, not just syntactic accuracy - Distinguish between response quality levels and articulate what separates adequate output from excellent output - Provide clear, opinionated written feedback on what works, what fails, and what feels off or misleading - Record short video walkthroughs explaining evaluations and reasoning - Help define quality standards for AI coding tools such as Cursor, Codex, and Claude Code

Requirements

- Senior, Staff, or Principal-level engineering background, or equivalent practical experience - Strong proficiency in TypeScript/JavaScript or Python - Hands-on experience with at least one modern AI coding assistant (OpenAI Codex, Claude Code, or Cursor) - Deep familiarity with AI-assisted development workflows and engineering best practices - Ability to evaluate code quality without executing it or reviewing every line in detail - Strong written and spoken English (B2 or above), since the role requires detailed feedback and video recordings - Comfort giving direct, opinionated feedback and defending subjective judgments

Nice to have

- Prior experience with prompt design or evaluation workflows - Background mentoring senior engineers or defining engineering standards

Benefits and work setup

- Fully remote, contract engagement open worldwide - Hourly rate of $100-$200 depending on experience and location - 10-20 hours per week with flexible scheduling - Project-based work running from roughly two weeks to a few months each, with new projects offered to evaluators who perform well - Single take-home evaluation exercise with a recorded video walkthrough serves as the hiring process; no traditional interview required