Kaushalendra Pratap Singh

AI Engineer

Demand Forecasting System

Time-series prediction for inventory planning

Problem

Inventory planning in retail lacks predictive accuracy, leading to either stockouts during high-demand periods or excess inventory during slow seasons.

Approach

Built a time-series forecasting system using Prophet that models trend, seasonality, and holiday effects on historical sales data. The system decomposes the signal to provide interpretable predictions.

Outcome

Delivered actionable forecasts supporting inventory planning and demand optimization.

Tech

PythonProphetMLflowPandasFastAPI

More Projects

Back to all projects