Join Mercor as an AI Evaluation Expert, where you'll author original research problems for cutting-edge AI models. This part time role invites PhD-level scientists to leverage their expertise and creativity.
Mercor is seeking experts in mathematics or applied mathematics for a six-week engagement. You will utilize your knowledge in subdomains of numerical linear algebra, computational mechanics, or computational finance. Your primary task involves creating scientific evaluation tasks that challenge today's AI frontier models and ensure robust performance assessment.
Key Responsibilities:
• Source materials like papers or datasets for task creation
• Write scientific prompts based on sourced materials
• Develop grading criteria for task evaluation




• Calibrate tasks against leading models for accuracy
• Ensure strong models fail more often than they succeed
Requirements:
• PhD in mathematics or a closely related field
• Expertise in at least two specified subdomains
• Proficient in Python for scientific computing
• Familiarity with Git and Docker workflows
• Strong analytical skills and problem-solving abilities
Utilize your mathematical expertise at Mercor, pushing the boundaries of AI evaluation through innovative problem creation.
#J-18808-Ljbffr

📌 AI Evaluation Expert at Mercor (Toronto)
🏢 Mercor
📍 Toronto