Data Scientist(ML & Predictive Analytics)

Role Overview We are looking for a Data Scientist with expertise in classical machine learning and predictive analytics to build custom models for equipment manufacturing and industrial environments. You will analyze sensor, time-series, and operational data to solve critical challenges, including predictive maintenance, equipment failure prediction, quality forecasting, and process optimization. Key Responsibilities Model Development: Design, train, validate, and deploy end-to-end classical ML and time-series models using Python. Feature Engineering & Analytics: Clean and transform structured, sensor, and time-series data from SCADA, PLC, MES, and ERP systems into actionable features. Problem Solving: Evaluate and optimize algorithms based on business objectives, metrics, and interpretability requirements. Cross-Functional Collaboration: Partner with manufacturing, process engineering, QA, and maintenance teams to translate operational bottlenecks into ML solutions. Deployment & Monitoring: Deploy production-ready models into operational workflows and continuously monitor performance and reliability. Technical Stack & Qualifications Mandatory Technical Skills Classical ML: Regression, Decision Trees, Random Forest, XGBoost, LightGBM, CatBoost, SVM, KNN, Clustering (K-Means, DBSCAN), Dimensionality Reduction (PCA), Anomaly Detection, and Ensemble Methods. Time Series Analysis: ARIMA, SARIMA, Prophet, and time-series forecasting techniques. Programming & Libraries: Python, SQL, pandas, NumPy, scikit-learn, SciPy, statsmodels, XGBoost, LightGBM, CatBoost, Matplotlib, Seaborn. Statistics & Data Prep: Hypothesis testing, regression analysis, experimental design, feature engineering, and robust ETL/preprocessing pipelines. Domain Experience & Qualifications Proven track record building and deploying ML models from scratch (AutoML experience alone is not sufficient). Hands-on experience handling structured industrial/sensor datasets. Strong communication and stakeholder management skills. Preferred & Nice-to-Have Manufacturing Domain: Exposure to Automotive, Heavy Engineering, Process Manufacturing, IIoT, OEE, Root Cause Analysis, or Downtime Reduction. Advanced Analytics: Predictive Maintenance, Remaining Useful Life (RUL) modeling, Survival Analysis, or Digital Twins. Deployment & MLOps: Docker, Cloud (AWS / Azure / GCP), Model Monitoring, Edge Analytics, or Streaming Data. Show more