Materials Science Task Architect

Mercor; United States; Remote; USD 90,000 - 130,000; Full time

Job summary

Mercor is seeking a senior contributor to design authentic AI evaluation tasks in materials science and related engineering domains. You will craft prompts, assemble data rooms, and establish objective grading criteria to measure model performance.

You will build the data room and scoring rubric, run tasks against models, analyze outcomes, and iterate until the evaluation clearly discriminates quality. Collaboration with domain experts is essential.

Qualifications

• Senior expertise in materials science or electrical/mechanical engineering.
• Ability to define tasks, data rooms, and grading criteria.
• Ability to explain why a solution is wrong, not just correct.

Responsibilities

• Author domain-authentic tasks drawn from real work.
• Build the data room and grading criteria that make the task objectively scorable.
• Run the task against the model, analyze where it succeeds or fails, and iterate.
• Collaborate with the project lead and domain experts to set benchmarks.

Skills

Materials science

Electrical engineering

Mechanical engineering

Critical thinking

Analytical thinking

Tools

CFD software

CAE tools

Job description

Mercor is seeking a senior contributor to design authentic AI evaluation tasks in materials science and related engineering domains. You will craft prompts, assemble data rooms, and establish objective grading criteria to measure model performance.

You will build the data room and scoring rubric, run tasks against models, analyze outcomes, and iterate until the evaluation clearly discriminates quality. Collaboration with domain experts is essential.