The Indian agricultural sector continues to face major inefficiencies due to the persistent mismatch between crop production and localized market demand. Farmers often cultivate crops without accurate insights into city-level requirements, resulting in oversupply, wastage, reduced profits, or undersupply leading to scarcity and inflation. To address this issue, the proposed system “Seasonal Crop Supply Chain Planner with Local Demand Forecasting” integrates artificial intelligence and web technologies to provide actionable demand forecasts for farmers. The system is designed as a web application where farmers can register with basic profile details. Upon login, farmers gain access to localized demand forecasts for key seasonal crops. Dataset downloaded from internet serve as the basis for model training. The forecasting module employs a Long Short-Term Memory (LSTM) network, which is capable of capturing complex temporal dependencies, seasonal variations, and non-linear patterns in agricultural demand data. The expected outcome of this system is to minimize post-harvest losses, improve decision-making for crop cultivation and supply, stabilize farmer incomes, and enhance food availability for consumers. Performance of the forecasting model will be evaluated using statistical measures such as Root Mean Square Error (RMSE). The system has the potential to serve as a decision-support tool for small and medium-scale farmers, contributing toward a more efficient and sustainable agricultural supply chain in India.