AI Low-Code Editor Prompt Segmentation for Targeted Property Updates

Solution Overview

Problem

Existing low code editors lack efficiency in content generation and resource utilization due to non-targeted prompts to AI models, leading to suboptimal performance and increased computational demands.

Innovation Solution

A low code editor that utilizes AI models to identify specific properties of a data instance and generate targeted new values through a two-stage interaction process, enhancing prompt specificity and reducing computational resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If non-targeted prompts are used to interact with AI models in low code editors, then the system can process general requests, but computational resources and network bandwidth are wasted due to suboptimal AI model responses

Engineering Contradiction:

Improvecontent generation efficiencyVSAvoidcomputational resource utilization

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The patent segments the data instance into multiple properties and uses targeted prompts to interact with AI models property-by-property rather than processing the entire data instance with a single general prompt. This segmentation allows the system to send more specific, efficient prompts that require fewer computational resources while maintaining high productivity in content generation.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent changes the parameter of prompt specificity by dynamically generating prompts that are tailored to each specific property of the data instance. Instead of using fixed, non-targeted prompts, the system adapts the prompt content to match the specific property being processed, thereby improving AI model response quality and reducing unnecessary computational resource consumption.

Inventive Principle:

Principle #35Parameter changes

2Productivity

If non-targeted prompts are used to interact with AI models in low code editors, then the system can process general requests, but network bandwidth is increased due to suboptimal AI model responses

Engineering Contradiction:

Improvecontent generation efficiencyVSAvoidnetwork bandwidth consumption

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The patent segments the data instance into multiple properties and uses targeted prompts to interact with AI models property-by-property rather than processing the entire data instance with a single general prompt. This segmentation allows the system to send more specific, efficient prompts that require fewer computational resources while maintaining high productivity in content generation.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent changes the parameter of prompt specificity by dynamically generating prompts that are tailored to each specific property of the data instance. Instead of using fixed, non-targeted prompts, the system adapts the prompt content to match the specific property being processed, thereby improving AI model response quality and reducing unnecessary computational resource consumption.

Inventive Principle:

Principle #35Parameter changes

3Adaptability or versatility

If AI models interact with entire data instances using non-specific prompts, then the system can handle complex requests, but the AI model responses are suboptimal and require additional computational iterations

Engineering Contradiction:

Improverequest handling capabilityVSAvoidcomputational iteration time

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent segments the data instance into multiple properties and uses targeted prompts to interact with AI models property-by-property rather than processing the entire data instance with a single general prompt. This segmentation allows the system to send more specific, efficient prompts that require fewer computational resources while maintaining high productivity in content generation.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary action by identifying and selecting the specific property that needs to be processed before interacting with the AI model. This preliminary selection ensures that each AI model interaction is focused and targeted, reducing the need for additional iterative computations and improving overall system efficiency.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20260079676A1Low code editor that uses ai model(s)

Publication Date: 2026.03.19 SALESFORCE INC

AI summary

A low code editor allows for a user to input (e.g., natural language text expressing a request to modify the visual representation). A first stage of interactions with at least one of a set of AI models is used to identify a property of a part of a data instance from which the visual representation is rendered. A second stage of interactions with at least one of the AI models is used to generate a new value for the identified property. In some implementations, these interactions include provision of some metadata regarding the identified property from a schema.