New opportunity

Senior Data Scientist

NKT · Chennai, India

About this role

Role Purpose: As a Senior Data Scientist at NKT, you will join an innovation team focused on applying AI and machine learning to real-world manufacturing challenges. You will work within a global organization to improve factory operations, product quality, and predictive maintenance, transforming complex industrial data into actionable insights. Responsibilities: You will collaborate with cross-functional teams across engineering, IT, and business to deliver practical outcomes. Key tasks include analyzing large-scale manufacturing datasets to identify patterns and anomalies, building machine learning models for predictive maintenance and defect detection, and developing solutions for process optimization and demand forecasting. You will also be responsible for communicating insights to both technical and non-technical stakeholders. Qualifications: Candidates should hold a Master’s or PhD in Data Science, Computer Science, Statistics, Engineering, or a related field. Required experience includes applying data science in manufacturing or industrial environments. Proficiency in Python, R, SQL, and machine learning libraries such as scikit-learn, TensorFlow, and PyTorch is essential. Experience with data visualization tools like Power BI or Tableau and cloud platforms such as Azure Fabric is required. Professional Attributes: The ideal candidate is detail-oriented, curious, collaborative, and capable of managing shifting priorities. Benefits: NKT offers remote work possibilities, flexible working hours, a strong safety culture, and career development opportunities within a high-tech international environment.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

PythonRSQLscikit-learnTensorFlowPyTorchPower BITableauAzure FabricPredictive MaintenanceDemand ForecastingProcess Optimisation

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

Industrial Data Application
The core requirement is to translate complex manufacturing datasets into actionable insights. You must demonstrate the ability to apply machine learning techniques specifically to industrial use cases like predictive maintenance, defect detection, and process optimization.
Cross-Functional Collaboration
Success depends on working effectively with engineering, IT, and business stakeholders. You must be able to translate technical model outputs into clear, non-technical business value to ensure initiatives are successfully implemented across the organization.
Technical Proficiency in Manufacturing Stack
You must be highly proficient in Python, R, and SQL, alongside deep learning frameworks like TensorFlow or PyTorch. Experience with cloud-based data environments, specifically Azure Fabric, is essential for managing large-scale industrial data pipelines.

Likely challenges to think through

Bridging Technical Models and Operational Reality
Manufacturing environments often involve noisy, high-volume data that may not be perfectly labeled. You may face difficulty in ensuring that theoretical machine learning models translate into reliable, real-world operational improvements on the factory floor.
Managing Shifting Priorities in a Global Context
Working within a global organization with multiple production sites means balancing competing demands from different regional stakeholders. You should prepare to discuss how you maintain focus and project momentum when business requirements or technical constraints change rapidly.

Practice your answers

Technical practice

Predictive Maintenance Modeling

What to coverDescribe your approach to building a predictive maintenance model for industrial equipment. Focus on how you handle sensor data, feature engineering for time-series anomalies, and the selection of appropriate algorithms to minimize false negatives.

Manufacturing Data Challenges

What to coverManufacturing datasets often contain noise, missing values, or imbalanced classes. Explain your strategy for cleaning and preprocessing large-scale industrial data to ensure model reliability and performance.

Demand Forecasting Techniques

What to coverCompare different statistical and machine learning approaches for demand forecasting in a manufacturing context. Discuss how you account for seasonality, external market factors, and the trade-offs between model complexity and interpretability.

Cloud-Based Data Pipelines

What to coverGiven the requirement for Azure Fabric, explain how you would design an end-to-end data pipeline that ingests raw manufacturing data and serves it for real-time or batch machine learning inference.

Model Evaluation in Production

What to coverHow do you evaluate the success of a process optimization model beyond standard metrics like RMSE or accuracy? Discuss the importance of business KPIs and how you validate that a model provides actionable, real-world improvements.

Deep Learning Frameworks

What to coverWhen choosing between TensorFlow and PyTorch for a specific industrial use case, what factors do you consider? Provide an example of how you have utilized one of these libraries to solve a complex pattern recognition problem.

Data Visualization for Stakeholders

What to coverDescribe how you use Power BI or Tableau to translate complex model outputs into insights for non-technical manufacturing managers. Focus on how you ensure the visualization drives decision-making rather than just displaying data.

SQL for Industrial Analytics

What to coverExplain how you optimize complex SQL queries when working with massive manufacturing datasets. Discuss your experience with indexing, partitioning, or window functions to handle high-volume time-series data efficiently.

Behavioural practice

Cross-Functional Collaboration

What to coverDescribe a situation where you had to collaborate with engineering or IT teams to implement a data-driven solution. How did you align on technical requirements and ensure the final output met the needs of both groups?

Managing Shifting Priorities

What to coverIn a fast-paced industrial environment, priorities can change rapidly. Provide an example of how you managed competing demands while maintaining the quality and progress of your data science projects.

Communicating Technical Insights

What to coverTell me about a time you had to explain a complex machine learning model to a non-technical stakeholder. How did you adjust your communication style to ensure they understood the business impact and limitations of your work?

Curiosity and Continuous Learning

What to coverThe field of AI is constantly evolving. Give an example of a new technique or tool you recently learned and how you applied it to improve a project or solve a challenging problem.

Sharpen your knowledge

Topics to prepare

Predictive Maintenance and Anomaly Detection
Review statistical and machine learning approaches for identifying equipment failure patterns in industrial time-series data. Exercise: Design a conceptual framework for detecting anomalies in sensor data streams, focusing on how to distinguish between normal operational noise and genuine equipment degradation.
Industrial Process Optimization
Understand the application of AI to improve manufacturing efficiency and quality control. Exercise: Research common optimization challenges in cable manufacturing or similar high-tech production environments and map them to specific machine learning techniques like regression or reinforcement learning.

Skills to revise

Azure Fabric for Data Analytics
Gain familiarity with the Microsoft Azure Fabric ecosystem for end-to-end data integration and analytics. Exercise: Explore the architecture of Azure Fabric to understand how it handles large-scale manufacturing datasets and integrates with machine learning workflows.
Machine Learning Model Implementation
Demonstrate proficiency in building and deploying models using Python libraries such as scikit-learn, TensorFlow, or PyTorch. Exercise: Build a prototype model using a public manufacturing dataset to perform a binary classification task, such as predicting a defect versus a non-defect.
Data Visualization for Stakeholders
Develop the ability to translate complex model outputs into actionable insights using Power BI or Tableau. Exercise: Create a dashboard mockup that visualizes key performance indicators for factory operations, ensuring the design is accessible to both technical and non-technical business stakeholders.

Your preparation checklist

Prepare a manufacturing-focused project walkthrough
Select a past project where you applied machine learning to industrial or manufacturing data. Prepare to explain the specific problem (e.g., predictive maintenance or defect detection), the data preprocessing steps for large-scale datasets, the model selection rationale, and the measurable operational impact.
Review your experience with Azure Fabric and cloud workflows
Be ready to discuss your experience with cloud-based data platforms. Specifically, prepare examples of how you have utilized Azure Fabric or similar cloud environments to manage, process, and deploy machine learning models within an enterprise architecture.
Develop a communication strategy for non-technical stakeholders
Prepare a concise explanation of a complex model or insight you delivered to a non-technical audience. Focus on how you translated technical metrics (like precision/recall or model drift) into business value, such as cost savings or improved factory efficiency.
Formulate questions about NKT's data infrastructure
Prepare thoughtful questions regarding the current state of data maturity at NKT. Ask about the integration of AI into existing factory operations, the primary challenges in data collection from industrial equipment, or how the innovation team balances experimental research with production-level deployment.
Practice explaining predictive maintenance and forecasting models
Be prepared to discuss the technical trade-offs in building models for predictive maintenance versus demand forecasting. Review your proficiency in Python, R, and libraries like scikit-learn, TensorFlow, or PyTorch, and be ready to justify your choice of tools for specific industrial use cases.

Availability last checked: 18 Oct 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in India. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…