Forecasting Data Scientist Bangalore Urban (India)

Job Title: Demand Forecasting Data Scientist

Experience: 6+ years

Location: BENGALURU, India (Full time)

If interested please share me your resume,

Job Summary

We are seeking an experienced Demand Forecasting Data Scientist with robust expertise in time series forecasting, machine learning, and supply chain analytics . The ideal candidate will have hands-on experience working with Databricks and Python

Key Responsibilities

Design, develop, and deploy demand forecasting models using time series, machine learning, and deep learning techniques.

Build end-to-end forecasting pipelines on Databricks leveraging distributed computing and scalable architectures.

Apply statistical and ML techniques such as ARIMA/SARIMA, Prophet, XGBoost, Random Forest, LSTM, GRU , TimeGPT and hybrid models.

Evaluate and improve forecast accuracy using appropriate metrics (MAPE, wMAPE, Bias).

Contribute to best practices in coding, model governance, documentation, and MLOps workflows.

Required Skills & Qualifications

Technical Skills

6 years of experience as a Data Scientist with a robust focus on demand forecasting or supply chain analytics .

Strong hands-on experience with Python and its associated libraries

NumPy, Pandas, SciPy

scikit-learn

statsmodels

TensorFlow / PyTorch (for DL models)

Solid experience with Time Series Forecasting techniques and real-world demand planning use cases.

Hands-on experience with Databricks , including Spark, notebooks, and distributed data processing.

Practical exposure to Machine Learning and Deep Learning model development and deployment.

Experience working with large, complex datasets in cloud-based or big data settings.

Practical understanding of Nixtla library is a plus.

Experience with MLOps , model monitoring, and CI/CD pipelines.

Domain Knowledge

Strong Supply Chain domain knowledge , especially demand forecasting, inventory planning, and sales forecasting.

Understanding of forecast lifecycle, demand variability, and business constraints. .