Data Scientist

• This role is for a Data Scientist in the Supply Chain & Manufacturing domain, with specific experience in areas like retail supply chain, inventory planning, and forecasting. The team is focused on predictive analytics and machine analytics, building real forecasting and ML models rather than just operating existing tools. GenAI and LLMs are being explored, but they are not mature yet and not the main focus of the role today.

• The team is currently working with KumoAI (recommendation software). They are building systems around inventory forecasting and vendor recommendation, and want to take those solutions further. Inventory planning is a key business problem they’ve already solved once and are now enhancing with more advanced ML and deep learning approaches.

• They are looking for someone who has hands-on experience building different types of forecasting models and machine learning models in Python, not someone who only consumes models or uses black-box solutions. The expectation is that this person has designed their own models from scratch and productionized them using platforms like AWS and Databricks. They’ve even built internal software for forecasting that the business can directly use.

• From a technical perspective, they want strong experience in deep learning for time series forecasting, especially for inventory forecasting and planning. There is very little NLP involved, so this is not an NLP-heavy role. The profile leans more toward a Full Stack Data Engineer / Data Scientist hybrid — someone who understands data pipelines, modeling, and deployment end-to-end.

• They specifically do not want black-box model users. They need someone who understands what is happening inside the models, can tune and improve them, and apply them directly to real business problems.