Mathematics Expert - AI Training

Braintrust; Colombia; Remote; COP 43,348,000 - 130,045,000; Part time

Job summary

Braintrust seeks a PhD-level mathematician to help improve large language models by producing rigorous, transparent, step-by-step solutions and analyzing model limitations. You will collaborate with AI researchers to align task designs with evaluation goals, including symbolic manipulation and multi-step reasoning.

The engagement runs 12 weeks, requires 20–40 hours per week with four hours overlap in Pacific Time, and pays per task at $80 USD for approved work.

Qualifications

• PhD in Mathematics or equivalent technical field is required.
• Strong mathematics knowledge spanning advanced concepts from graduate-level to PhD.
• Excellent English comprehension and structured written communication.
• Ability to provide constructive feedback remotely and work independently in a fast-paced setting.

Responsibilities

• Create high-quality, step-by-step solutions with detailed reasoning.
• Collaborate with AI researchers to align task designs with evaluation goals (symbolic manipulation, abstraction, multi-step reasoning).
• Construct novel scenarios and evaluate complex reasoning pathways.
• Provide structured feedback and detailed annotations to improve model quality.

Skills

Analytical thinking

Problem-solving

Structured writing

English proficiency

Education

PhD in Mathematics

Job description

Use advanced mathematics and analytical reasoning to help improve large language models. You will probe model limitations and explain complex concepts through clear, step-by-step reasoning.

What You Will Do

• Create high-quality, step-by-step solutions with detailed reasoning.
• Work with AI researchers to align task designs with evaluation goals, including symbolic manipulation, abstraction and multi-step reasoning.
• Construct novel scenarios and evaluate complex reasoning pathways.
• Provide structured feedback and detailed annotations to improve model quality.

What You Will Bring

• A PhD in Mathematics or an equivalent technical field.
• Strong mathematics knowledge spanning advanced engineering-entrance through graduate and PhD-level concepts.
• Strong analytical, research and problem-solving skills with excellent English comprehension.
• Exceptional structured written communication and the ability to provide constructive feedback remotely.
• Lateral thinking and the ability to work independently in a fast-paced remote setting.
• A personal desktop or laptop and stable high-speed internet.

Experience in AI evaluation, data annotation, content review, quality assurance or a related analytical role is preferred, not required.

PROJECT DETAILS

Expected engagement length is 12 weeks, with an immediate start. A minimum of 20 hours per week, up to 40 hours per week, is required. Four hours of daily Pacific Time overlap is required. Selection includes an assessment and a delivery review.

PAYMENT

Turing pays $80 USD directly for each approved task.