Senior AI Coding Evaluator (Remote) — Engineering Taste

Work From Anywhere; United Kingdom; Remote; GBP 104,000 - 208,000; Part time

Job summary

Work From Anywhere is seeking a highly experienced software engineer (SR+) to evaluate AI coding agents such as OpenAI Codex and Claude Code. This is not a traditional coding role; you will judge whether the model thinks like a great engineer and if responses are useful and well-reasoned.

The position is remote worldwide, 10-20 hours per week, with take-home tasks and Loom walkthroughs. Applicants should be senior-level, hands-on with Codex/Claude Code/Cursor, and able to produce clear written

Qualifications

• Senior, Staff or Principal-level software engineer.
• Hands-on with Codex, Claude Code or Cursor.
• Strong written English and ability to provide clear feedback.
• Experience with AI-assisted dev workflows.

Responsibilities

• Evaluate AI-generated coding interactions end to end.
• Provide clear, opinionated feedback on usefulness and accuracy.
• Assess explanations and reasoning beyond code.
• Help define criteria of excellence with tools like Codex/Cursor.

Skills

Engineering judgment

Feedback writing

English proficiency

Subjective evaluation

Video explanations

High standards

Tools

OpenAI Codex

Claude Code

Cursor

Job description

Work From Anywhere is seeking a highly experienced software engineer (SR+) to evaluate AI coding agents such as OpenAI Codex and Claude Code. This is not a traditional coding role; you will judge whether the model thinks like a great engineer and if responses are useful and well-reasoned.

The position is remote worldwide, 10-20 hours per week, with take-home tasks and Loom walkthroughs. Applicants should be senior-level, hands-on with Codex/Claude Code/Cursor, and able to produce clear written