Kaggle-Expert Data Scientist (Remote, Flexible Hours)

Weekday AI; United States; Remote; USD 83,000 - 165,000; Full time

Job summary

Weekday AI is seeking a competitive data scientist with Kaggle experience to design and evaluate challenging data science tasks for AI coding systems. You will create reproducible environments, verifiable reference solutions, and robust evaluation methods to measure correctness and quality.

The role emphasizes rigorous reasoning, experimentation with approaches, debugging, and evaluating AI agents across competitions. Flexible, fully remote engagement with ~15 hours per week.

Responsibilities

• Task Development: Design challenging data science, ML, and modeling tasks based on real-world problems and competition experience.
• Environment Setup: Build reproducible execution environments, prepare datasets, configure dependencies, ensure consistent task execution.
• Reference Solutions: Implement accurate, efficient reference solutions for each task.
• Testing and Evaluation: Develop automated tests, validation criteria, and evaluation frameworks to distinguish correct solutions.
• AI Agent Evaluation: Run AI coding agents against designed tasks, analyze performance, refine task specs.
• Debugging and Quality Assurance: Identify bugs, data leakage, and methodological weaknesses impacting quality.
• Technical Documentation: Document problem statements, outputs, methodologies, and verification procedures.
• Continuous Improvement: Improve task complexity, reproducibility, and evaluation reliability through iteration.

Skills

Competitive Data Science

Kaggle Competitions

Machine Learning

Python Programming

Statistical Analysis

Data Modeling

Job description

Weekday AI is seeking a competitive data scientist with Kaggle experience to design and evaluate challenging data science tasks for AI coding systems. You will create reproducible environments, verifiable reference solutions, and robust evaluation methods to measure correctness and quality.

The role emphasizes rigorous reasoning, experimentation with approaches, debugging, and evaluating AI agents across competitions. Flexible, fully remote engagement with ~15 hours per week.