LLM API Connectors for Resilient Cross-System Request Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional API connectors are brittle, costly, and difficult to maintain due to frequent API changes and service interruptions, requiring robust, scalable, and flexible solutions.
Innovation Solution
An AI-based connector system using large language models (LLMs) to interpret and translate API requests, enabling modular and secure integration across disparate systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional programming is used to map endpoints and values between APIs, then connectivity between systems is achieved, but the system becomes highly sensitive to symbol and formatting changes, causing working code to fail inexplicably
Solution Approach 1:
The patent replaces traditional mechanical symbol-matching programming with a semantic understanding system using large language models. Instead of rigid string comparison and format matching, the LLM interprets the meaning and context of API parameters, enabling robust connectivity that adapts to formatting changes while maintaining reliability.
Solution Approach 2:
The system changes the parameter representation from fixed symbolic strings to semantic descriptions generated by LLMs. This transformation allows the connector to understand API parameters in terms of their meaning rather than their exact symbolic representation, making the system resilient to parameter name changes and formatting variations.
2Adaptability or versatility
If companies build custom libraries to connect APIs, then specific integration requirements are met, but the connectors become brittle and difficult to maintain due to frequent API changes
Solution Approach 1:
The patent creates a universal LLM-based connector framework that can handle multiple API integrations through a single system. The LLM's semantic understanding capability allows it to adapt to different API schemas and requirements without requiring custom programmed mappings for each integration, reducing maintenance burden while maintaining versatility.
Solution Approach 2:
The LLM automatically generates the mapping logic between source and destination APIs by understanding the semantic meaning of parameters. This self-service capability eliminates the need for manual programming and frequent updates when APIs change, as the LLM can autonomously adapt to new schemas and requirements.
3Reliability
If traditional API connectors are built to handle frequent changes, then connectivity is maintained, but the cost of building and running these connectors increases significantly
Solution Approach 1:
The patent replaces expensive traditional programming approaches with LLM-based semantic processing. The LLM's ability to understand and interpret API parameters reduces the need for extensive custom development and testing, thereby lowering development costs while maintaining reliable connectivity across changing APIs.
Data Source
AI summary
A method for API connectors includes receiving from a user account a first API request formatted for a source system and comprising an intent, selecting a first action corresponding to the intent, generating a first prompt based on the first action, providing the first prompt to a first large language model (LLM) and receiving a response having a structured description of the intent, using the structured description of the intent to select a second action, generating a second prompt based on the second action, providing the second prompt to a second LLM and receiving a response having a second API request formatted for a destination system, providing the second API request to the destination system, receiving from the destination system a response to the second API request, and providing the response to the second API request to the user account.