Data Scientist (Python & SQL) - Freelance AI Trainer

Role overview

Mindrift brings domain experts into project-based AI work for major technology companies, with an emphasis on testing, evaluating, and enhancing AI systems. This is not a permanent role; participation happens on a project basis.

What you may work on

Each assignment can differ, but typical contributor work may include designing original computational data science tasks that mirror real analytical work across sectors such as telecom, finance, public services, e-commerce, and healthcare.

• Build Python-based problems that require libraries such as Pandas, NumPy, SciPy, scikit-learn, Statsmodels, Matplotlib, and Seaborn.
• Create tasks that are computationally demanding and cannot be solved manually in a practical amount of time.
• Develop exercises that require multi-step reasoning in data preparation, statistical analysis, feature engineering, predictive modelling, and insight generation.
• Keep tasks fully deterministic and reproducible by avoiding randomness or using fixed random seeds where needed.
• Ground problems in realistic business use cases such as customer analytics, risk analysis, fraud detection, forecasting, optimisation, and operational improvement.
• Design end-to-end scenarios that cover the whole data science workflow, from data ingestion and cleaning to EDA, modelling, validation, and deployment considerations.
• Include large-scale data processing situations that call for efficient and scalable methods.
• Validate solutions using Python along with standard data science tools and statistical techniques.
• Write clear problem statements with practical business context and provide checked, correct answers.

What the client is looking for

This opportunity suits data science professionals with Python experience who are open to part-time, non-permanent project work.

• At least 5 years of practical data science experience with clear business outcomes.
• A portfolio of delivered projects and publications that demonstrate applied problem-solving.
• Advanced Python skills for data science work, including pandas, numpy, scipy, scikit-learn, and statsmodels.
• Strong command of statistical analysis and machine learning, including both methods and practical use cases.
• Advanced SQL skills and experience working with databases for manipulation and analysis.
• Exposure to GenAI tools such as LLMs, RAG, prompt engineering, and vector databases.
• Understanding of MLOps processes and model deployment workflows.
• Familiarity with modern frameworks such as TensorFlow, PyTorch, and LangChain.
• Strong written English at C1 level or above.

How the work process works

The flow is: apply, complete the qualification stage, join a project, complete assigned tasks, and receive payment.

Time commitment

During active project phases, the estimated workload is around 10 to 20 hours per week. This is only an estimate and is not guaranteed.

Compensation

Compensation for this project can reach the equivalent of up to $58 per hour, depending on expertise level and pace of contribution. Pay can differ from one project to another based on scope, complexity, and the skills required.

Application note

Submit your CV in English and include your English proficiency level.