LLM Shopping List Generation Under User Budget Constraints

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing online concierge systems lack a scalable mechanism to automatically build shopping carts based on user-defined budget constraints using conversational search queries.

Innovation Solution

Utilize a language model to generate a list of items constrained by user-defined constraints, such as budget, by prompting a large language model (LLM) with customer queries and item information, and present the results through a user interface.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If manual item selection and cart building is used, then users can control their shopping choices, but the process is inefficient and requires repetitive manual adjustments

Engineering Contradiction:

Improveshopping cart building efficiencyVSAvoidtime spent on repetitive manual adjustments

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system enables automatic cart building by having the system serve itself through AI-generated shopping lists that automatically populate cart items based on user constraints, eliminating the need for manual item-by-item selection and repetitive adjustments

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The manual mechanical process of selecting and adding items to cart is replaced by an AI-based automated system that generates shopping lists and populates carts through computational processes, substituting human manual operations with intelligent automation

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If automated cart building is implemented, then shopping efficiency is improved, but scalability to large user populations is challenging

Engineering Contradiction:

Improveautomatic shopping list generationVSAvoidsystem scalability

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The automated cart building process is segmented into distinct modular components: constraint extraction from user queries, AI model inference for item selection, and cart population execution. This segmentation enables independent optimization and scaling of each component to handle large user populations

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The AI-based shopping list generation system is designed as a universal platform that can serve multiple users simultaneously with different constraints and preferences, handling diverse shopping scenarios through a single scalable architecture rather than user-specific custom solutions

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Extent of automation

If AI models are used to generate shopping lists, then automated constraint-based item selection is achieved, but computational resources are consumed

Engineering Contradiction:

Improveconstraint-based item selection automationVSAvoidcomputational resource consumption

Core Design Contradiction:

Extent of automationVSUse of energy by moving object

Solution Approach 1:

The system applies partial automation by using AI models only for the critical decision-making aspects (item selection based on constraints) while relying on deterministic processes for routine operations (cart population, constraint application), reducing overall computational resource consumption compared to full AI-based processing

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS12585678B2Using language model to automatically generate list of items at an online system based on a constraint

Publication Date: 2026.03.24 MAPLEBEAR INC

AI summary

Embodiments relate to using a large language model (LLM) to generate a list of items at an online system with a user defined constraint. The online system receives a query that includes at least one constraint. The online system generates a prompt for input into the LLM, based at least in part on the query. The online system requests the LLM to generate, based on the prompt, a set of constraints for a set of item types. The online system generates a list of candidate items by searching through a set of items stored in one or more non-transitory computer-readable media using the set of constraints for the set of item types. The online system causes a device of the user to display a user interface with the list of items for inclusion into a cart, the list of items obtained from the list of candidate items.