Data Scientist for ML Forecasting - #7412685

HP invites applications for a motivated and talented Data Scientist to join our AI for Inventory & Product placement team, nested within the GTM Advanced Analytics organization.

In this important role, you will employ implement advanced statistical and/or machine learning methodologies to maximize the forecasting accuracy in our models supporting global sales and inventory forecasts for thousands of products, distributors and resellers weekly. You will be involved in the design, development, training and validation of state of the art data science models (DeepAr, Transformers, Tree Based). You will be the central connection among Data Scientists, Data Engineers, MlOps and Stakeholders, ensuring that business needs are aligned with team efforts and focused on objectives.

Responsibilities:

• Advanced Machine Learning Design
  • Use R or Python to analyse and clean extensive datasets, ensuring actionable insights are extracted to optimize our algorithms.
  • Analyse and integrate various factors impacting inventory and supply chain decisions into data science models. (pricing, promotions, inflation, etc..)
  • Identify the best suited forecasting signals, as well as algorithms/models suited to the problem
    
• Machine Learning Forecasting
  • Ensure there is no leakage, and address signals drift in the forecasting pipelines
  • Design, develop and validate the selected forecasting, backtesting and inference pipeline
    
• Engineering and forecasting inference
  • Work with MlOps and engineers to ensure the scalability and efficiency of the forecasting pipeline
  • Monitor accuracy at scale across thousands of products and distributors
  • Run anomaly detection techniques to identify edge cases and areas of improvement
  • Run A/B testing to replace and deploy new ML models with higher accuracy and more robust to pattern changes

Basic Qualifications:

• 2-3 years working experience in Data Analytics or Data Science.
• PhD degree in Business Analytics, Statistics or related field is a must.
• Experience with Pyspark is a must.
• Experience with Pytorch or Keras is a must.
• Experience in Supply Chain, Inventory Management or Distribution networks is a strong plus.
• Experience with Git and cloud-based notebooks are a plus.
• Proficiency in SQL, PowerBI is a plus.
• Technical Proficiency:
  • Strong proficiency with the design, development, and validation of advanced machine learning or statistical forecasting techniques in Python.
  • Strong knowledge regarding various industry standard and state of the art algorithms and models used in forecasting (DeepAr, Transformers, Neural Networks, Tree based algorithms, etc..)
    
• Communication Skills: Capability to extract requirements from users and adeptly communicate intricate data insights to data analysts, data scientists and data engineers.