Natural-Language Workflow Generation with Editable Placeholders

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Generating and modifying workflows using traditional tools is tedious and time-consuming, especially in low-code or no-code environments, as they require building workflows activity-by-activity with graphical user interfaces, leading to inefficiencies and potential human errors.

Innovation Solution

A workflow generation tool that uses large language models (LLMs) to generate skeleton workflows based on natural language inputs, allowing users to modify placeholder activities and generate complete workflows or playbooks with graphical user interfaces.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If traditional workflow generation tools are used to build workflows activity-by-activity with graphical user interfaces, then workflows can be created with detailed control over each activity, but the process becomes tedious and time-consuming

Engineering Contradiction:

Improveworkflow generation accuracyVSAvoidworkflow creation time

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent replaces the manual mechanical process of building workflows activity-by-activity through graphical user interfaces with an automated language model-based system. The language model generates workflow code directly from natural language descriptions, eliminating the need for manual drag-and-drop operations and significantly reducing workflow creation time while maintaining accuracy through programmatic generation.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The workflow generation system performs self-service by automatically generating workflow code from natural language descriptions without requiring manual intervention for each activity. The language model autonomously creates the workflow structure, activities, and connections, freeing users from tedious manual configuration while preserving the ability to review and modify the generated workflows.

Inventive Principle:

Principle #25Self-service

2Ease of operation

If manual workflow creation through graphical user interfaces is used, then users can visually design workflows, but computational resources and processor utilization are unnecessarily consumed

Engineering Contradiction:

Improveworkflow design accessibilityVSAvoidcomputational resource consumption

Core Design Contradiction:

Ease of operationVSUse of energy by moving object

Solution Approach 1:

The patent substitutes the computationally intensive graphical user interface rendering and manipulation with a language model-based code generation approach. Instead of rendering visual elements and handling drag-and-drop operations that consume significant processor resources, the system generates workflow code directly from text, dramatically reducing computational resource consumption while maintaining ease of operation through natural language input.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Adaptability or versatility

If traditional workflow tools are used, then comprehensive workflow properties can be specified, but the process is prone to human errors

Engineering Contradiction:

Improveworkflow configuration flexibilityVSAvoidworkflow creation accuracy

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent replaces manual workflow configuration with automated code generation from language models. This substitution eliminates human errors associated with manual property specification while preserving comprehensive workflow configuration flexibility. The language model accurately translates natural language descriptions into precise workflow code, ensuring reliability without sacrificing the ability to specify detailed workflow properties.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system incorporates feedback mechanisms where the generated workflow code can be reviewed, validated, and modified by users. This feedback loop ensures that the automated generation process maintains adaptability and versatility, allowing users to verify that the generated workflow accurately reflects their intentions while reducing errors through automated validation and correction capabilities.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20250272639A1Systems and methods for generating workflows based on natural language inputs using large language models

Publication Date: 2025.08.28 SERVICENOW INC

AI summary

A method includes receiving a natural language request to generate a workflow, where the natural language request specifies at least one characteristic of the workflow, generating, using one or more large language models (LLMs), a skeleton workflow based on the at least one characteristic, where the skeleton workflow includes first and second placeholder activities, and where the first placeholder activity comprises a first placeholder value for a first property of the first placeholder activity receiving an input requesting to modify the skeleton workflow, and updating the skeleton workflow based on the input.