Join a cross-functional Data Science team delivering cutting-edge machine learning solutions in the field of time series forecasting. You will be part of a long-term initiative aimed at building scalable and high-quality ML-driven applications for a global client from the [finance/pharma/tech – choose depending on client sector. The project emphasizes model robustness, modular software design, and continuous integration/deployment of ML pipelines.
responsibilities :
Design, develop, and maintain machine learning models using Python.
Implement and evaluate supervised learning models and hybrid approaches.
Work with time series data and contribute to forecasting solution development.
Collaborate with software engineers to transform prototype models into production-ready software packages.
Apply object-oriented programming principles to structure scalable codebases.
Contribute to the design and execution of model evaluation processes (e.g., validation sets, overfitting mitigation).
Participate in CI/CD processes, unit testing, and ML pipeline deployment.
Cooperate closely with data scientists, DevOps engineers, and domain experts.
requirements-expected :
3+ years of professional experience using Python in ML solution development.
Solid understanding of machine learning techniques and algorithms (e.g., supervised learning, train/test splits, overfitting, evaluation metrics).
Ability to write clean, efficient, and modular Python code.
Hands-on experience with object-oriented programming (OOP).
Basic understanding of time series analysis principles.
Experience in structuring ML code into maintainable software packages.
Knowledge of or willingness to learn time series forecasting techniques.
Familiarity with Deep Learning frameworks applied to forecasting (e.g., TensorFlow, PyTorch, GluonTS, Prophet).
Exposure to ML engineering tools and practices (e.g., CI/CD pipelines, unit testing, MLflow, Airflow).