Fully Remote and Freelance Project- 8-10 hours per week expected as a minimum

Opportunity for experienced data scientists to evaluate how frontier AI models handle real engineering work:

About the Role:

In this role, you will handle real data science work:

• An analysis from messy data to a defensible conclusion,
• An experiment designed and read out correctly,
• A model built and validated without leakage,
• SQL and Python that actually does what the ticket asked.
• You bring the judgment you have built sending back a notebook that passed its cells but not its logic, and knowing the difference between a result and a real result. We bring the model output that judgment is needed to grade.

You will design challenging, realistic tasks drawn from your own practice, such as:

• An exploratory analysis and write up from a messy dataset,
• An A/B test design and read out,
• Afeature set and model with validation,
• A SQL pipeline with data quality checks,
• A metric definition and dashboard spec,
• A forecast with error analysis, run them through frontier AI agents, and evaluate what comes back against a professional standard.
• You will work with realistic professional files, the kind a practitioner in your field actually handles,which you assemble yourself. Some tasks are compact, built around a handful of files; othersare larger scenarios that take several days to build. In every case the goal is the same: a task a competent professional in your field would complete correctly and a frontier model currently gets wrong.

Qualifications Required:

• Professional applied data science experience preferred (analytics, machine learning, statistical modeling, experimentation, analysis adjacent data engineering) in industry or research. Undergraduate study does not count toward the experience floor.
• In progress Bachelor’s degree or higher in a quantitative field, completed or in progress.
• Advanced degree is a plus but optional; practical applied work outweighs credentials; cloud and analytics certifications are neither required nor screened for.
• Working proficiency in Python and/or SQL. You write, debug and can explain your own analysis code; reading and writing code is part of the work.
• Depth in at least one of: product or business analytics and BI; applied machine learning; statistics and experimentation (A/B testing, causal inference); forecasting; analysis adjacent data engineering; applied NLP or computer vision with shipped work.
• Has independently owned a multi step analysis end to end, from raw or messy data through cleaning, modeling and interpretation to a decision, rather than picking up a clean dataset mid pipeline.
• Can judge whether a result is real and say what it means: leakage, confounding, selection effects, multiple testing, power, and interpreting results in business or scientific term

This is not a traditional data science role. You will be helping build better AI by putting your

knowledge to work in a structured, flexible, fully remote environment. The work is long form and

self directed, and clear written reasoning matters as much as technical depth.

The work is long form and self directed, and clear written reasoning matters as much as technical depth

Process:

Skills assessment: about 2-4 hours, unpaid

(domain assessment plus a short training project)

For more information and how to sign up, please get in touch