Data Science & Analysis - AI Training - San Jose, US

prolificacademicltd; United States; Remote; USD 91,000 - 129,000; Full time

Benefits offered by this job

Remote work

Job summary

Prolific Academic Ltd is seeking a data science professional to contribute to AI training and evaluation through project-based tasks. The role focuses on assessing statistical reasoning, reviewing model-generated data insights, and checking code outputs.

Candidates complete a skills assessment before assignments; projects vary in duration and may require an uninterrupted hour. Remote and flexible schedule offered, with compensation up to $80 per hour depending on task and demand.

Qualifications

• Bachelor’s, master’s or doctoral qualification in data science, statistics, mathematics or a closely related quantitative discipline.
• Real world experience in data analysis, predictive modeling or machine learning engineering.
• Strong statistical problem solving skills and ability to interpret complex data visualizations independently.
• High attention to detail and ability to identify subtle errors in statistical arguments or interpretations.
• Extensive use of large language models in data workflows, with awareness of common failure modes.
• Ability to explain technical concepts clearly; a PayPal account is required to receive payment.

Responsibilities

• Review AI produced data insights and statistical analyses for accuracy, clarity and completeness.
• Check mathematical proofs, statistical formulas and data conclusions against reliable references.
• Run data scripts in Python or R and confirm their results, including behavior on edge cases.
• Record where models perform well or poorly on data problems and predictive modeling.
• Verify that model responses meet task instructions and professional standards.

Skills

Statistical problem solving

Data interpretation

Explain technical concepts

Attention to detail

Experience with LLMs

Education

Bachelor’s/Master’s/PhD in data science or related

Real world data analysis/ML experience

Tools

Python

R

SQL

Tableau/Power BI

Matplotlib/Seaborn

PyTorch/TensorFlow

Job description

Role overview

Use professional data science or analytics experience to contribute to AI training and evaluation through project based tasks. The work centers on assessing statistical reasoning, reviewing model generated data insights and checking code outputs. A short skills assessment is required before candidates become eligible for assignments, which vary in duration and can require an uninterrupted hour.

Responsibilities

• Review AI produced data insights and statistical analyses for accuracy, clarity and completeness.
• Check mathematical proofs, statistical formulas and data conclusions against reliable references.
• Run data scripts in Python or R and confirm their results, including behavior on edge cases.
• Record where models perform well or poorly on data problems and predictive modeling.
• Verify that model responses meet task instructions and professional standards.

Requirements

• Bachelor’s, master’s or doctoral qualification in data science, statistics, mathematics or a closely related quantitative discipline.
• Real world experience in data analysis, predictive modeling or machine learning engineering.
• Strong statistical problem solving skills and ability to interpret complex data visualizations independently.
• High attention to detail and ability to identify subtle errors in statistical arguments or interpretations.
• Extensive use of large language models in data workflows, with awareness of common failure modes.
• Ability to explain technical concepts clearly; a PayPal account is required to receive payment.

Nice to have

• Python skills with Pandas, NumPy or scikit-learn, or R; advanced SQL; visualization experience with Tableau, Power BI, Matplotlib or Seaborn; and knowledge of PyTorch, TensorFlow or statistical modeling.

Benefits and work setup

Work is remote and flexible, with task availability depending on research demand. The source describes project rates of up to $80 per hour, though actual rates and task lengths vary.