Benchmarking & Evaluation Researcher

Home/Careers/Benchmarking & Evaluation Researcher

Core AI Research

Benchmarking & Evaluation Researcher

This team builds standardized tests and benchmarks to evaluate AI models.

Apply for this roleManage an applicationBack to careers

Overview

This team builds standardized tests and benchmarks to evaluate AI models.

Responsibilities

• Develop benchmark datasets: Create datasets and tasks for evaluating accuracy, fairness, robustness, and safety.
• Design evaluation protocols: Define evaluation metrics, conditions, and processes to measure model performance.
• Analyze results: Interpret outcomes and provide feedback to improve models.
• Collaborate with external organizations: Work with consortiums to develop open benchmarks for transparency.
  
  

Qualifications

• Master's or doctoral degree in statistics, data science, or related field.
• Experience in dataset creation, statistical analysis, and evaluation frameworks.
  
  

Equal opportunity

TAIRC evaluates candidates using role-related evidence and documented review criteria. Employment decisions must not be based on protected personal characteristics.

Engagement pathways and compensation status

TAIRC is currently building its funding base. Opportunities identified as volunteer or unpaid do not provide compensation at this stage. Any future paid appointment would require available funding, organizational approval, applicable legal compliance, and a separate written agreement. Participation does not guarantee future employment or compensation.

• Volunteer contribution interest may be reviewed only as a voluntary, scoped pathway.
• Educational internship interest requires separate supervision, educational purpose, and legal review.
• Advisory expressions of interest are not board or formal advisory appointments.
• Future compensated employment is contingent on funding, approval, legal compliance, and a separate written offer.
  
  

Accessibility and accommodations

For an accommodation or alternate submission format, contact [email protected]. The application form does not request medical documentation or sensitive identity documents.

Posting information

Department Core AI Research

Public role ID TAIRC-ROLE-015

Status Open

Location, employment type, workplace type, compensation, sponsorship policy, posted date, and closing date are omitted unless HR has entered approved factual values.

Recruitment privacy

Application data is used for recruitment review, stored with restricted access, and handled under the recruitment privacy and retention policy.