Data Scientist

We're looking for a Data Scientist to build production-grade models on large-scale industrial and railway/metro telemetry data. You'll work across time-series forecasting, anomaly detection, predictive maintenance, and Applied GenAI, helping engineering and control-room teams detect failures earlier and make operational knowledge easier to access.

Responsibilities:

• Build forecasting models for energy, traction power, load profiles, and operational planning.
• Develop anomaly detection for multivariate sensor and telemetry streams.
• Build predictive maintenance and Remaining Useful Life (RUL) models.
• Work with real-world industrial data, including missing/stale tags, sensor drift, irregular sampling, and limited failure labels.
• Apply signal processing techniques such as spectral analysis, wavelets, and feature extraction.
• Build RAG systems over maintenance records, incident reports, O& M manuals, and engineering standards.
• Use LLMs/GenAI for structured extraction from work orders, fault logs, and operational reports.
• Prototype text-to-SQL / natural-language interfaces for operational data.
• Design evaluation frameworks covering accuracy, grounding, refusal behavior, and reliability.
• Deploy and monitor models in production, including drift detection and retraining.

Requirements:

• 4+ years of applied data science experience with models deployed in production.
• Strong expertise in time series, including ARIMA/SARIMAX, ETS, state-space models, gradient boosting, and deep learning approaches such as LSTM/GRU, TCN, or transformers.
• Experience with forecasting and anomaly detection on sensor/telemetry data.
• Hands-on GenAI/LLM application development: RAG, embeddings, vector databases, prompt engineering, structured outputs, and evaluation.
• Strong Python and SQL skills.
• Proficiency with pandas, NumPy, scikit-learn, and PyTorch/TensorFlow.
• Strong statistical fundamentals, temporal validation, leakage prevention, and uncertainty quantification.
• Experience with MLOps and tools such as MLflow, Weights and Biases, or similar.
• Ability to collaborate closely with domain/engineering teams and translate domain knowledge into modelling features and constraints.

Good to Have:

• Experience with industrial/sensor data.
• Railway or metro domain experience.
• Predictive maintenance, reliability engineering, RCM/FMEA/RAMS/survival analysis.
• Experience with time-series foundation models such as Chronos, TimesFM, or Moirai.
• LLM fine-tuning, agent frameworks, guardrails, or safety tooling.
• Explainability and experience deploying models for operational users.
• Experience working in regulated engineering environments.