Data Scientist

Location: Brussels, Belgium

Company: Tata Consultancy Services (TCS) Belgium

Employment Type: Full-time

Nature of the Tasks

• Collaborate with stakeholders to collect requirements, frame business problems as data science hypotheses, define success metrics, and develop or deploy advanced data mining and machine learning solutions.
• Collaborate with User Experience (UX) and product teams to specify design requirements for the effective presentation and interpretation of model outputs and insights.
• Develop process to monitor and analyse data accuracy.
• Identify, collect, and prepare data for analysis, collaborating with Data Analysts to ensure robust production-grade data pipelines, with a focus on feature engineering and data readiness for modelling.
• Produce data models according to specific problem statements.
• Develop and implement machine learning algorithms, statistical models, and scripts to solve specific business problems.
• Collaborate with Data Analysts and Architects on the design of the analytics architecture to ensure it supports the scalability and performance requirements of data science models.
• Write the different documentation associated with the tasks and liaise with other teams as necessary to address cross-system interdependencies.
• Develop visualizations to communicate model behaviour, key insights, and performance metrics, and collaborate with Data Analysts on integrated dashboard reporting.
• Design, develop, and evaluate predictive models capable of handling both structured and unstructured data, including model selection, training, and hyperparameter tuning.
• Elaborate processes to ensure the compliant implementation of regulatory frameworks (e.g., EU AI Act), including model risk management, monitoring for drift/bias, and comprehensive documentation (e.g., model cards).
• Ensure that model development, deployment, and monitoring practices adhere to all relevant regulatory frameworks and internal governance policies.
• Communicate model insights, limitations, and behaviour effectively to both technical and non-technical audiences, and create interpretable outputs and documentation.
• Conduct rigorous experiment design (e.g., A/B testing), employ cross-validation techniques, and perform statistical significance testing to validate findings and model performance.
• Deploy and maintain models into production using Machine Learning Operations (MLOps) practices, including Continuous Integration/Continuous Deployment (CI/CD pipelines, model registries, and ensuring reproducibility.

Specific Expertise and Technologies

• Knowledge of advanced analytics techniques and tools (e.g., Python, R, SAS, Spark) to design and implement data-driven solutions.
• Knowledge of machine learning and natural language processing (e.g., Scikit-learn, TensorFlow, PyTorch, Hugging Face) to build predictive and text-based models.
• Knowledge of programming languages (e.g., Python, R, SQL; Perl optional) commonly used in data science for modelling and automation.
• Knowledge of MLOps practices (e.g., CI/CD pipelines, model registry, unit testing frameworks) to ensure reproducibility and quality in model delivery.
• Knowledge of business intelligence tools (e.g., Tableau, SAS, SAP Analytics) to visualise and communicate model outputs and insights.
• Knowledge of data engineering and ETL processes using tools (e.g., Talend, Informatica, dbt, Azure Data Factory) to prepare training datasets.
• Knowledge of data storage and query technologies (e.g., SQL, NoSQL, MongoDB, Hadoop) to extract and process data at scale.
• Knowledge of designing scalable data storage solutions (e.g., data lakes, lakehouses, distributed stores) for advanced analytics and AI workloads.
• Knowledge of advanced analytics applications such as forecasting, recommendation systems, anomaly detection, or sentiment analysis for business impact.
• Knowledge of AI governance principles (e.g., transparency, explainability, fairness, accountability) aligned with emerging EU AI Act requirements.
• Knowledge of AI compliance, risks, and mitigation practices (e.g., bias monitoring, drift detection, model cards).
• Knowledge of data and AI regulatory frameworks (e.g., GDPR, EU AI Act) and ability to design models compliant with legal and ethical standards.
• Knowledge of experiment design (e.g., A/B testing, cross-validation, significance testing) to validate models rigorously.
• Knowledge of model deployment practices (e.g., containerization, APIs, serving frameworks) to operationalise machine learning solutions.

Minimum Level of Expertise

• Normal

Certification and/or Standards

Optio