Data Scientist

Looking for a Data Science professional with expertise in PySpark/Databricks and experience working on different stages of a Data Science project life cycle. Incumbent is expected to build and optimize Data Pipelines, tune/enhance models, and explain output to business stakeholders. Team primarily works on Demand Forecasting, Promotion Modelling, and Inventory Optimization problems for CPG/Retail customers, prior experience in CPG/Retail is strongly preferred.

Role & responsibilities

• Design, optimize, and maintain scalable ETL pipelines using PySpark and Databricks on cloud platforms (Azure/GCP).
• Develop automated data validation process to proactively perform data quality checks.
• Optimal allocation of cloud resources to control cloud cost.
• Create and schedule jobs on Databricks platform. Work with GitHub repositories and ensure that best practices are being implemented.
• Work on Supply Chain Optimization models, such as Demand Forecasting, Price Elasticity of Demand, and Inventory Optimization.
• Build and tune forecast model, identify improvement opportunities, and perform experiments to prove value.
• Incumbent is expected to have frequent conversation with Business Stakeholders. Explain data deficiencies, forecast variances, and role of different forecast drivers.
• Follow best practices in Architecture, Coding, and BAU operations.
• Collaborate with cross-functional teams, such as Business Stakeholders, Engagement Managers, Data Ops/Job Monitoring, Product Engineering/UI, Data Science, and Data Engineering.

Preferred candidate profile

• Preferred Education: bachelors in computer science/IT or in similar field with strong programming exposure and masters degree in Statistics/Operations Research/Mathematics/Data Science.
• 2 4 years of experience in Data Science/Data Engineering. Exposure to Demand Forecasting and Inventory optimization in CPG/Retail will be a big plus.
• Proven experience building and optimizing Data Pipelines/ETL processes in PySpark, DataBricks, Python (Pandas/NumPy) and SQL. Experience working with Git as a collaboration tool.
• Good understanding of cloud platform – preferably Azure.
• Exposure to conventional time series forecasting (ESM, ARIMA) and Machine Learning models (GBM, ANN, Random Forests).
• The role is expected to work independently with very low supervision.
• Good communication skills, ability to present output to business stakeholder and convey data deficiencies.