Role description

As a Data Scientist in our prognostics team, you work within a defined scope — contributing to the analytics pipeline across two tracks: component health and prognostics models, and analytical foundations for uptime-related services. You own your models end-to-end, grow your technical depth, and are expected to deliver, learn fast, and raise the quality bar of what you ship.

Build & Deliver Analytical Models

• Develop predictive and prognostic models for component and system-level health monitoring — covering failure prediction, wear estimation, and anomaly detection.
• Build and validate models using vehicle telematics, sensor data, maintenance records, and usage patterns.
• Implement vehicle-level health indicators: degradation scoring, RUL (Remaining Useful Life) estimation, and threshold-based or ML-driven alerting.
• Contribute to prescriptive logic — translating model outputs into actionable maintenance recommendations.
• Develop analytical intelligence for uptime-related services — including usage-based insights, fleet-level health patterns, and data foundations for new service offerings beyond component prognostics.
• Ensure your work is robust, well-tested, interpretable, and production-ready.

Contribute to the Analytics Pipeline

• Work closely with prognostics engineers to understand domain requirements and translate them into analytical solutions.
• Participate in early feasibility assessments — helping scope what is analytically possible before full development begins.
• Contribute to requirements definition for data needs, model inputs, and expected outputs.
• Hand off analytical logic clearly to IT and industrialization teams, ensuring continuity from model to deployed solution.

Grow the Practice

• Apply and promote best practices in experimentation, validation, and model governance within the team.
• Document your work rigorously — assumptions, methods, limitations, and results.
• Share knowledge actively and contribute to the team's collective analytical maturity.
• Explore new approaches and bring fresh ideas into the roadmap — through structured experimentation and curiosity.

Who You Are

Mindset

You are delivery-oriented and intellectually curious. You take ownership of your models from first exploration to production handoff. You are rigorous in your work, ask the right questions when requirements are unclear, and care about the real-world impact of what you build. You are comfortable working in an engineering environment where domain knowledge matters as much as modelling skill.

Technical Qualifications

• Master's degree or higher in Computer Science, Statistics, Mathematics, Engineering, or equivalent.
• Experience in data science for several years in industrial applications
• Background in mechanical or electrical engineering fundamentals.
• Solid foundations in ML and statistical modelling: regression, classification, time-series, anomaly detection.
• Good software discipline: Python, SQL, version control (Git), clean and documented code.
• Familiarity with data manipulation and analysis libraries (pandas, scikit-learn, etc.).
• Experience or strong interest in big data and cloud environments (Databricks, Spark, Azure).
• Fluent English.
• Interest in how analytics creates value in a service business context — uptime, maintenance contracts, or fleet operations.

Nice to Have

• Knowledge of OBD/CAN data, ECU logs, or telematics data architectures.
• Experience with survival models, deep learning for time-series (LSTM, Transformers), or unsupervised methods.
• Familiarity with MLOps practices: model versioning, CI/CD, monitoring.
• Curiosity about uptime and aftermarket service business models — understanding what a service offering is and how analytics can enable it.