Computational Mathematics Problem Author and Reviewer
Use advanced mathematics and Python to author and review research-grade problems that help train large language models. This remote, three-month contractor assignment offers 20, 30, or 40 hours per week.
Generative AI & RLHF
100% Remote
Worldwide
Eligibility
Entry
Experience
Aug 12, 2026
Posted
Open worldwide
Interested in this role?
Create a free OpenTrain account and apply in minutes.
OpenTrain AI is the hiring and contracting organization for this role and the #1 platform for finding and building careers in AI training and data labeling. OpenTrain helps specialists discover meaningful projects, build a credible AI training portfolio, and grow their work in a rapidly expanding industry.
Remote work with a global contributor community
A profile that helps showcase your AI training experience
Opportunities to apply advanced subject-matter expertise to cutting-edge AI development
About AI Training Work
AI training is the human side of building modern artificial intelligence. Specialists create examples, assess model responses, and review reasoning so that large language models can become more accurate, reliable, and useful.
In this role, your mathematical expertise will support the development and evaluation of challenging computational problems and solutions used for model training.
Contribute directly to the development of advanced AI systems
Apply your academic and technical knowledge to flexible remote project work
Help improve model reasoning through rigorous problem and solution evaluation
The Role
OpenTrain AI is seeking a Computational Mathematics AI Problem Author and Reviewer to create challenging scientific problems and computational solutions for large language model training. You will work with research-grade and graduate-level quantitative problems, use Python to develop solutions, and apply rigorous standards for accuracy, reliability, and logical coherence.
The assignment combines original content creation with structured quality review. It is designed for a specialist who can formulate difficult mathematical tasks, assess proposed reasoning, and communicate decisions clearly.
Remote contractor assignment
Part-time commitment options of 20, 30, or 40 hours per week
Three-month contract with potential extension based on performance and project needs
English-language work
Worldwide eligibility
What You'll Do
You will develop, solve, and review computational mathematics content according to defined quality standards. The work requires careful attention to mathematical accuracy, problem difficulty, solution logic, and clear documentation.
Develop original research-grade or graduate-level computational mathematics problems using Python
Define problem categories, secondary tags, and difficulty levels
Solve formulated problems with accurate and logically coherent solution paths
Conduct structured reviews of problem and solution accuracy
Evaluate mathematical difficulty and logical coherence
Incorporate reviewer and team-lead feedback
Refine content and document decisions clearly
Collaborate proactively with reviewers
Requirements
This assignment requires advanced academic preparation in computational mathematics and demonstrated research experience. Candidates should be able to design demanding quantitative problems and produce or assess rigorous computational solutions.
PhD or PhD candidacy in Computational Mathematics
At least one published peer-reviewed paper in a scientific or mathematical domain
Advanced Python programming for scientific coding
Ability to design graduate-level quantitative problems
Ability to develop rigorous solution paths
Ability to evaluate mathematical accuracy, difficulty, and logical coherence
Strong analytical and problem-solving ability
Clear written and verbal communication skills
Who Should Apply
This opportunity may suit computational mathematics researchers, doctoral candidates, and quantitative specialists who enjoy both original problem design and detailed technical review. It is especially relevant for candidates who want to apply research-level expertise to the development of advanced language models.
Computational mathematics PhD candidates and graduates
Researchers with peer-reviewed scientific or mathematical publications
Python users with experience in scientific or computational coding
Analytical reviewers who can explain decisions and improve work through feedback
How the Assignment Works
You will contribute as a remote contractor for an initial three-month period. Choose a commitment of 20, 30, or 40 hours per week, with possible extension depending on performance and project needs.
Create your OpenTrain profile and apply through OpenTrain to be considered for this computational mathematics AI training assignment.
Contractor and part-time engagement
20, 30, or 40 hours per week
Three-month initial duration
Potential extension based on performance and project needs
Create and solve research-grade computational mathematics problems, validate Python solutions, and evaluate AI training materials. This part-time contract role requires PhD-level expertise and offers 20+ hours per week.
Create and solve original graduate-level computational mathematics problems in Python to train and evaluate frontier AI systems. This fully remote contract requires advanced scientific coding, rigorous review, and 20 or more hours per week.
Use advanced expertise in holography, AdS/CFT, and mathematical physics to review and adjudicate AI-generated solutions. Work remotely for 10 hours weekly over 8 to 10 weeks at $80 to $160 per hour.