AI Workflow Builder Automating Binary Code Generation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current AI system building methods are complex and performance-focused, lacking comprehensive solutions for implementing AI systems efficiently, as existing frameworks are designed for research and not optimized for real-world environments, and manual development is slow and cumbersome.

Innovation Solution

A system workflow builder that receives an AI system scenario definition, parses it to create an application workflow graph, maps it to an execution pipeline, and automatically generates executable binary code, with iterative optimization based on runtime performance to achieve performance targets.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If deep learning frameworks are used for AI system building, then AI capability is provided, but performance optimization is lacking as they are designed for research prototypes

Engineering Contradiction:

ImproveAI capabilityVSAvoidperformance

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

The system segments the AI system building process into distinct components: scenario definition, workflow graph generation, execution pipeline mapping, and binary code generation. This segmentation allows each component to be optimized independently while maintaining overall system capability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system changes parameters by automatically optimizing workflow configurations, mapping parameters to execution pipelines, and adjusting runtime parameters based on performance feedback. This transforms research-oriented frameworks into performance-optimized systems through parameter tuning and configuration optimization.

Inventive Principle:

Principle #35Parameter changes

2Ease of manufacture

If fixed AI streaming application structure SDKs are used, then a structured approach is provided, but additional development is required and performance optimization for real environments is lacking

Engineering Contradiction:

Improvestructured approachVSAvoidperformance optimization

Core Design Contradiction:

Ease of manufactureVSProductivity

Solution Approach 1:

The system performs preliminary actions by pre-defining execution pipelines, pre-mapping workflows to optimized structures, and pre-configuring performance parameters before runtime. This eliminates the need for additional development while providing both structure and performance optimization.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms that collect runtime performance data and use it to automatically adjust and optimize the execution pipeline. This closed-loop feedback enables continuous performance improvement without requiring manual re-development.

Inventive Principle:

Principle #23Feedback

3Adaptability or versatility

If manual design and development is performed from scratch, then complete system control is achieved, but the process becomes slow and complex

Engineering Contradiction:

Improvesystem controlVSAvoiddevelopment time

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system uses copying by generating workflow graphs from scenario definitions, copying optimized execution pipeline templates, and replicating proven configurations. This maintains complete system control while dramatically reducing development time through template-based generation.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system implements self-service by automatically generating executable binary code from workflow graphs without manual intervention. The system serves itself by performing scenario definition parsing, workflow generation, pipeline mapping, and code compilation autonomously, eliminating manual development steps while maintaining full control.

Inventive Principle:

Principle #25Self-service

4Adaptability or versatility

If API-based SDKs are used for system implementation, then existing structures can be leveraged, but additional development work is required

Engineering Contradiction:

Improveexisting structure utilizationVSAvoiddevelopment complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system replaces the mechanical API-based development approach with an automated compilation system that directly generates executable binary code from workflow graphs. This substitution eliminates the need for manual API integration and reduces development complexity by automating the implementation process.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11710027B2Artificial intelligence workflow builder

Publication Date: 2023.07.25 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

In some examples, a method includes receiving an artificial intelligence (AI) system scenario definition file from a user, parsing the definition file and building an application workflow graph for the AI system, and mapping the application workflow graph to an execution pipeline. In some examples, the method further includes automatically generating, from the workflow graph, application executable binary code implementing the AI system, and outputting the application executable binary code to the user. In some examples, the execution pipeline includes one or more building blocks, and the method then further includes collecting running performance of each of the building blocks of the execution pipeline in a runtime environment.