← Retour aux Emplois
Code Evaluation
$100 - $200 de l'heure

Senior Software Engineer, AI Training - Canada

Postuler Maintenant → →

Description

Evaluate engineering judgment in

AI coding Assess how coding agents such as OpenAI Codex, Claude Code and Cursor respond to realistic development situations. This work centers on reviewing interactions and engineering judgment, including whether explanations help a developer and whether the response earns their trust.

Evaluation responsibilities

  • Review coding interactions from beginning to end for usefulness, overall correctness and sound engineering decisions.
  • Assess preambles, explanations and reasoning alongside the generated code.
  • Distinguish levels of response quality and explain why one attempt deserves a different rating from another.
  • Write direct, specific feedback about what worked, what was misleading and what felt wrong to an experienced engineer.
  • Help define strong interactions with modern coding assistants, including how well they guide users. You should be able to make subjective but rigorous judgments and explain them clearly. The role does not involve writing production code, and it requires assessing code without executing it or inspecting every line.

Required experience Senior, Staff or

Principal engineering experience, or an equivalent background, is required. You need a strong foundation in TypeScript/JavaScript or Python, practical experience with at least one of OpenAI Codex, Claude Code or Cursor, and deep familiarity with development workflows that use AI assistants. Written and spoken English must be B2 or above. Deliverables include detailed written feedback and short recorded video explanations, so clear communication is essential. Experience designing prompts or evaluation workflows is helpful. Mentoring senior engineers or setting engineering standards is also useful, alongside a willingness to give direct feedback and maintain a high quality bar.

Rate, schedule and selection

The contract is remote and open worldwide, with flexible work of 10 to 20 hours per week. Pay ranges from $100 to $200 per hour depending on experience and location. The engagement is ongoing, with individual projects lasting about two weeks to a few months and additional projects offered to successful evaluators. Starting depends on passing the evaluation exercise and an available project opening. Selection consists of one evaluation exercise completed independently, accompanied by a recorded Loom walkthrough. There is no interview.

Details

Entreprise qui recrute

G2i

Category

Code Evaluation

Lieu

Toronto

Languages Required

🇺🇸 English

Publié

Senior Software Engineer, AI Training - Canada
$100 - $200 de l'heure
Postuler Maintenant → →