STEM Researcher - AI Trainer

Mercor; Madrid; Remote; EUR 25,000 - 41,000; Full time

Job summary

Mercor in Madrid seeks experienced scientists and engineers fluent in Spanish to train and evaluate AI agents on realistic tasks. You will work in digital workspaces with documents, spreadsheets, email, chat and industry tools to create and review tasks, define success criteria, and audit AI outputs for accuracy and local rules.

Collaborators will assess work quality, ensure compliance with local regulations, and provide detailed feedback.

Qualifications

• Experience as a scientist or engineer in research or industry.
• Native or fluent in Spanish.
• Bachelor's degree from a recognized institution in Spain.
• Comfortable with quantitative analysis, technical documentation and data-heavy problem solving.

Responsibilities

• Building workspaces: reviewing files, memos and records a real team would have for a project or case.
• Writing tasks: drafting realistic requests in your language that a colleague or client might send, grounded in local practice, regulations and terminology.
• Defining what good looks like: writing clear grading criteria and checking whether an AI agent's work would meet your professional standard.
• Reviewing AI work: going through an agent's step-by-step work, spotting mistakes such as wrong calculations, missed details or incorrect local rules, and correcting them.

Skills

Scientific background

Engineering experience

Quantitative analysis

Technical documentation

Education

Bachelor's degree in Spain

Master's degree or PhD a plus

Job description

We're looking for experienced scientists and engineers who are native or fluent in Spanish to help train and evaluate AI agents on realistic professional tasks.

What you'll do:

You'll work in realistic digital workspaces, with documents, spreadsheets, email, chat and industry tools, to create and review the kinds of tasks you handle in your own job. Your work will include:

• Building workspaces: reviewing files, memos and records a real team would have for a project or case.
• Writing tasks: drafting realistic requests in your language that a colleague or client might send, grounded in local practice, regulations and terminology.
• Defining what good looks like: writing clear grading criteria and checking whether an AI agent's work would meet your professional standard.
• Reviewing AI work: going through an agent's step-by-step work, spotting mistakes such as wrong calculations, missed details or incorrect local rules, and correcting them.

Requirements:

• Experience as a scientist or engineer in research or industry (master's or PhD a plus)
• Native or fluent in Spanish
• Bachelor's degree from a recognized institution in Spain
• Comfortable with quantitative analysis, technical documentation and data-heavy problem solving

Details: Remote, about 21 hours per week, hourly, pilot running Oct 23 to Nov 19.