STEM Data Annotation Specialist Remote Canada English C1, French C1 CAD 45 to 75 per hour

Corpshore Solutions Corporation; Canada; Hybrid; CAD 40,000 - 65,000; Part time

Benefits offered by this job

Fully remote

Job summary

Corpshore Solutions Corporation offers a fully remote, part-time opportunity across Canada for evaluating model reasoning in STEM disciplines. You will write problems, assess step-by-step solutions, and correct solutions with a clear rationale, while also reviewing software engineering code for correctness and efficiency.

The role targets graduate students, postdocs, engineers, and educators seeking substantive part-time work with flexible hours (minimum twenty hours weekly).

Qualifications

• A degree in a STEM discipline, with graduate level study or equivalent depth.
• Genuine subject expertise in at least one STEM field (math, physics, chemistry, biology, CS, engineering).
• Ability to work with LaTeX or comparable notation.

Responsibilities

• Write problems and evaluate model reasoning step by step.
• Correct and rewrite model-generated solutions with clear rationale for changes.
• Assess code correctness, efficiency and edge case handling in software programmes.
• Verify claims against authoritative sources.
• Document ambiguity and contribute to guideline refinement.

Skills

Analytical thinking

Mathematical reasoning

LaTeX proficiency

Academic research depth

Education

STEM degree

Graduate study or equivalent depth

Tools

LaTeX

Job description

Frontier models are now measured substantially on their reasoning in mathematics, physics, chemistry, biology, engineering and computer science. Improving that requires people who can tell a correct derivation from a plausible-looking wrong one, which is a much narrower population than general annotation requires and is paid accordingly.

You will write problems, evaluate model reasoning step by step, and correct solutions in your discipline. Fully remote across Canada, minimum twenty hours weekly. This work suits graduate students, postdoctoral researchers, working engineers and educators who want substantive part-time work.

What you will do

Evaluate model reasoning in your discipline, identifying not only whether the answer is wrong but where in the chain the reasoning failed, which is the part that actually improves a model. Author original problems and reference solutions at specified difficulty levels. Correct and rewrite model-generated solutions with a clear rationale for each change. Assess code correctness, efficiency and edge case handling on software engineering programmes. Verify claims against authoritative sources rather than intuition. Document ambiguity and contribute to guideline refinement where the current version produces inconsistent outcomes.

What you bring

A degree in a STEM discipline, and in most programmes graduate level study or equivalent professional depth. Genuine subject expertise in at least one of: pure or applied mathematics, statistics, physics, chemistry, biology, computer science, software engineering, electrical, mechanical, civil or chemical engineering. Ability to work with LaTeX or comparable notation. English at C1. Rigour, and the intellectual honesty to mark something wrong when you are not certain rather than waving it through.

Nice to have

Graduate degree or doctorate. Teaching, tutoring, marking or peer review experience. Publication record. Competitive mathematics or programming background. French at C1, which opens bilingual STEM programmes that pay at a premium. Professional engineering designation.

What we offer

Rates well above general annotation, reflecting the scarcity of the expertise. Fully remote and genuinely flexible, which works alongside graduate study or a full-time technical role. Paid guideline training. Work that materially and measurably improves model reasoning in your field. Routes into senior review and programme design.