BasketIQ | Devpost

BasketIQ Architecture
BasketIQ-Smart Planner-1
BasketIQ-SmartPlanner-2
BasketIQ-Dashboard-1
BasketIQ-Dashboard-2
clickhouse-dashboard-1
nimble-agents-usage-tracking
clickhouse-dashboard-2
nimble-agents
BasketIQ-Insights
BasketIQ-Upload Receipt
BasketIQ-ReceiptData
BasketIQ-PriceHunt-Agent

Inspiration

BasketIQ was inspired by the gap between spending visibility and spending intelligence. While existing budgeting apps show how much users spend on groceries, they do not explain what is being bought, how frequently items are purchased, or how to optimize future shopping decisions for cost savings. Grocery receipts contain valuable behavioral signals, but they are rarely structured or analyzed.

What it does

BasketIQ transforms raw grocery receipts into intelligent shopping insights. It analyzes recurring purchases, predicts consumption needs, detects inflation trends, and recommends where to buy each item across stores like Walmart, Sam’s Club, Amazon, and ShopRite to maximize savings through bulk buying and optimized store selection.

How we built it

We built BasketIQ using a multi-agent AI architecture. Google DeepMind is used for receipt understanding and reasoning, ClickHouse stores and analyzes long-term purchase history, and Nimbleway provides real-time pricing and product intelligence across multiple retailers.

Challenges we ran into

The biggest challenges were handling noisy and inconsistent receipt data, mapping items to standardized product categories, and combining historical behavior with real-time pricing data to generate accurate recommendations.

Accomplishments that we're proud of

We successfully built an end-to-end system that turns unstructured receipts into actionable grocery plans with bulk optimization, subscription detection, and store-level price recommendations.

What we learned

We learned how to design context-aware multi-agent systems, combine structured analytics with live web data, and build AI systems that move from insight generation to actionable decision-making.

What's next for BasketIQ

Next, we aim to add automated cart generation, deeper personalization per household member, and proactive purchasing where the system can auto-suggest or pre-build weekly grocery plans based on predicted consumption.