Remote | Competitive Data Scientist (Kaggle) — $60–$120/hour
About The Position
24-MAG LLC is seeking experienced data scientists, machine learning practitioners, and applied mathematics experts for a remote consulting opportunity focused on advanced AI training and evaluation. The work combines competitive data science experience with algorithmic reasoning, statistical analysis, programming, automated testing, and technical verification to improve AI systems’ reliability and problem-solving capabilities. The consultant will develop challenging data science and machine learning tasks, build reproducible computational environments, implement verified reference solutions, and evaluate AI coding agents. Assignments may involve identifying datasets and modelling objectives, validating implementations, assessing data leakage and methodological issues, and developing automated tests and objective evaluation criteria. Consultants will document task specifications, computational requirements, implementation decisions, mathematical reasoning, and verification procedures. This is a freelance/contract engagement, fully remote and open worldwide. The expected commitment is approximately 15 hours per week, with a flexible schedule. Compensation is $60–$120 per hour. Prior AI model training or evaluation project experience is not required.
Requirements
- Demonstrated experience in competitive data science, machine learning, quantitative modelling, or applied mathematical problem-solving.
- Participation in Kaggle, DrivenData, or comparable data science and machine learning competitions.
- Strong understanding of machine learning methodologies, statistical analysis, and mathematical reasoning.
- Practical programming skills sufficient to independently implement, test, and verify computational solutions.
- Ability to design technically challenging problems with clearly defined objectives and verifiable outcomes.
- Experience preparing datasets, configuring computational environments, and managing technical dependencies.
- Familiarity with reproducible experimentation and systematic validation of analytical results.
- Strong understanding of common modelling errors, flawed assumptions, and data leakage.
- Ability to evaluate algorithmic correctness and identify implementation weaknesses.
- Strong analytical reasoning, logical thinking, and independent problem-solving capabilities.
- Ability to investigate complex technical problems and explain underlying issues clearly.
- Strong written communication and technical documentation skills.
- Experience working independently on computational assignments and delivering reproducible technical outputs.
- Ability to manage technical assignments efficiently within a flexible remote consulting arrangement.
Nice To Haves
- Experience with competition platforms such as Zindi, AIcrowd, Numerai, DataSource.ai, or CodaLab/Codabench.
- Advanced Python programming experience, including debugging, testing, and working with unfamiliar codebases.
- Familiarity with machine learning engineering workflows and practical model evaluation techniques.
- Familiarity with AI coding agents or automated programming evaluation.
- Prior experience with AI model training or evaluation projects is not required.
Responsibilities
- Design technically challenging data science and machine learning tasks informed by real-world analytical problems and competitive data science environments.
- Develop clear task specifications with defined objectives, datasets, modelling goals, and computational requirements.
- Build reproducible computational environments with the required data, dependencies, and execution resources.
- Implement accurate reference solutions and validate them for correctness, reproducibility, and alignment with task requirements.
- Apply machine learning, statistical modelling, and mathematical reasoning to analytical challenges.
- Evaluate modelling strategies, algorithms, and computational techniques against specified objectives.
- Identify flawed assumptions, implementation errors, methodological inconsistencies, and potential data leakage.
- Develop automated tests and evaluation criteria to assess solution correctness.
- Run AI coding agents on assigned tasks and analyze their technical performance and errors.
- Refine task specifications, reference solutions, and verification procedures based on observed agent behavior.
- Document task objectives, computational requirements, implementation approaches, mathematical reasoning, and verification procedures.
- Review code, analytical outputs, and test results for correctness and consistency.
- Improve task clarity, computational reliability, and evaluation effectiveness while maintaining technical accuracy and reproducibility.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Career Level
Mid Level
Education Level
No Education Listed
