-
-
Home page of the dashboard
-
mode of the dashboard. It is used to enter the new stock load.
-
-
-
-
Inspiration
Inventory management is an important challenge for many businesses. Poor stock visibility, slow-moving products, and incorrect reorder decisions can lead to wasted resources and missed sales opportunities. We wanted to build a solution that makes inventory data easier to understand and helps users make smarter decisions using AI.
What it does
STOCK AI is an AI-powered inventory management and decision-support application built with Python and Streamlit. It provides an interactive dashboard for monitoring stock levels, incoming goods, sales, and inventory value.
The application includes inventory data entry, sales trend analysis, top- and slow-moving product insights, reorder recommendations, an AI assistant powered by NVIDIA Nemotron, and an audit trail for reviewing inventory activities.
Users can explore inventory information through charts and tables and ask questions in natural language to gain additional insights.
How we built it
We built STOCK AI using Python and Streamlit for the interactive web interface. We organized the inventory operations and data processing in separate Python modules and used a database to store and manage inventory records.
For AI-powered assistance, we integrated the NVIDIA Nemotron API so users can ask questions about their inventory and receive contextual responses. We also developed analysis views to help users understand product performance and identify potential stock replenishment needs.
The application is deployed using Streamlit Community Cloud, making it accessible through a web link.
Challenges we ran into
One challenge was integrating the AI assistant with the application and configuring the NVIDIA API credentials securely for cloud deployment. We also worked on organizing inventory data so it could support useful analysis and contextual AI responses.
Another challenge was preparing the application for deployment, managing dependencies, and ensuring that the interface and core features worked correctly in the hosted environment.
Accomplishments that we're proud of
We are proud to have developed a working inventory management application that combines traditional inventory operations with AI-powered assistance in a single interface.
The dashboard, product analysis, reorder support, and natural-language AI assistant demonstrate how AI can make inventory information more accessible and actionable. We are also proud to have deployed the application online so it can be demonstrated through a shareable link.
What we learned
Through this project, we learned how to integrate an AI API into a web application, structure a Python project into separate modules, visualize inventory data, manage application secrets, and deploy a Streamlit application to the cloud.
We also learned that building a useful AI application involves more than connecting a model: reliable data, clear user interactions, secure configuration, and deployment testing are equally important.
What's next for STOCK AI
Our next steps are to improve inventory forecasting, enhance reorder recommendations, strengthen persistent database storage, and add support for barcode scanning and automated inventory updates.
We also plan to improve the AI assistant's analytical capabilities and continue testing the application with realistic inventory scenarios to make STOCK AI more useful for businesses.
Log in or sign up for Devpost to join the conversation.