GAI Reuse Flowchains for Low-Code Enterprise LLM Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Integrating generative artificial intelligence (GAI) into enterprise platforms is a non-trivial task due to technical challenges and inefficiencies, including the need for developer skills, time, and resources to code specific prompts and data retrieval for each unique use case, leading to code duplication and maintenance issues.
Innovation Solution
A GAI reuse service utilizing scenario flowchains that enable time- and resource-efficient integration of GAI by providing a design-time tool for configuring and testing scenario flowchains, allowing applications to connect to a GAI reuse service via a URL, minimizing code duplication and streamlining integration processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If GAI is integrated into enterprise platforms using traditional coding methods, then functional capabilities are achieved, but development time and resource consumption increase significantly
Solution Approach 1:
The patent applies preliminary action by pre-configuring scenario flowchains with all necessary GAI integration parameters, data retrieval logic, and prompt templates during design-time. This allows the integration architecture to be established in advance, eliminating the need for time-consuming custom coding during deployment and significantly reducing development time while maintaining full functional capabilities.
Solution Approach 2:
The patent implements copying by creating reusable scenario flowchain templates that can be instantiated multiple times across different applications. Instead of coding GAI integration from scratch for each use case, the same proven flowchain configurations can be copied and adapted, dramatically improving integration efficiency and reducing both development time and resource consumption.
2Adaptability or versatility
If custom GAI integration is coded for each unique use case, then specific functionality is achieved, but code duplication increases
Solution Approach 1:
The patent applies universality by designing scenario flowchains as multi-functional templates that can serve multiple different use cases through configuration rather than custom coding. A single flowchain template can be adapted to various GAI applications by changing parameters and data sources, providing use case flexibility while eliminating code duplication and reducing overall system complexity.
Solution Approach 2:
The patent implements segmentation by breaking down GAI integration into discrete, independently configurable scenario flowchain modules. Each flowchain represents a specific functional segment that can be developed, tested, and maintained separately. This modular approach allows different use cases to be built by combining and configuring these segments rather than writing duplicate custom code for each scenario.
3Reliability
If traditional GAI integration methods are used, then application functionality is achieved, but maintenance burden increases
Solution Approach 1:
The patent applies copying by using standardized scenario flowchain templates that can be replicated across multiple applications. When maintenance or updates are needed, changes can be made to the template once and automatically propagated to all instances, ensuring functional reliability while dramatically easing maintenance burden compared to custom-coded integrations that would require individual updates.
Solution Approach 2:
The patent implements preliminary action by establishing standardized flowchain configurations and best practices during design-time. This preliminary setup creates a proven, reliable integration pattern that can be consistently applied, reducing functional risks while making future maintenance easier since the architecture and procedures are already documented and tested in advance.
Data Source
AI summary
Methods, systems, and computer-readable storage media for receiving a scenario service request for a scenario of an application, determining a scenario flowchain represented in the scenario service request, retrieving a scenario flowchain configuration of the scenario flowchain from a database, the scenario flowchain configuration including a data object that defines a set of steps that are to be executed in an order, each step being associated with a step type and a set of parameters, executing steps in the set of steps, where at least one step is executed to prompt a large language model (LLM), and returning a result to the application, the result comprising a response from the LLM that is responsive to the prompt.