Granted Patent B2

US 12,613,790 · App. 18/489,774 · Granted Apr 28, 2026

Inventors: Levon Korganyan (Tarzana, CA); Meel Velliste (San Ramon, CA); Emrah Diril (Myrtle Beach, SC); Sergey Numerov (Walnut Creek, CA); Ashutosh Chaurasia (Rajasthan, IN); Shreyans Shrimal (Telangana, IN); Varun Dhall (Uttar Pradesh, IN); Ravi Sharda (Karnataka, IN)

Assignee: Fivetran Inc.

G06F11/3616G06F8/60G06F9/544G06F11/3006G06F16/2458G06F16/3344G06F16/93G06F16/951G06F40/35

Abstract

Claims (54)

1 . A computer-implemented method for monitoring connectors deployed in a data collection system, the method comprising:

deploying a data connector generated from a connector knowledge representation, the connector knowledge representation generated from an API source knowledge graph using a large language model;

receiving information describing a defect reported in connection with the deployed data connector;

determining a portion of the connector knowledge representation associated with the defect;

mapping the portion of the connector knowledge representation to a portion of the API source knowledge graph;

identifying a portion of an API documentation associated with the portion of the API source knowledge graph; and

reporting information describing the defect, the information comprising at least the portion of the API documentation associated with the portion of the API source knowledge graph.

2 . The method of claim 1 , wherein the API source knowledge graph comprises a plurality nodes, each node corresponding to a header of an API source document.

3 . The method of claim 2 , wherein a given API source document has a given header and has a reference to a different API source document having a different header.

4 . The method of claim 3 , wherein the API source knowledge graph comprises an edge between a first node corresponding to the given header and a second node corresponding to the different header based on the reference to the different API source document.

5 . The method of claim 1 , wherein determining the portion of the connector knowledge representation associated with the defect comprises:

determining an API feature that is causing the defect; and

identifying a node in the API source knowledge graph corresponding to the API feature.

6 . The method of claim 5 , wherein identifying the portion of the API documentation associated with the portion of the API source knowledge graph comprises:

determining a set of nodes depending from the node in the API source knowledge graph based on there being one or more directional edges pointing from the node to each node of the set of nodes; and

determining the portion to include the set of nodes.

7 . The method of claim 6 , wherein reporting the information describing the defect comprises:

outputting one or more links to API documentation that corresponds to each node of the set of nodes.

8 . A non-transitory computer-readable medium comprising memory with instructions encoded thereon for monitoring connectors deployed in a data collection system, the instructions, when executed, causing one or more processors to perform operations, the instructions comprising instructions to:

deploy a data connector generated from a connector knowledge representation, the connector knowledge representation generated from an API source knowledge graph using a large language model;

receive information describing a defect reported in connection with the deployed data connector;

determine a portion of the connector knowledge representation associated with the defect;

map the portion of the connector knowledge representation to a portion of the API source knowledge graph;

identify a portion of an API documentation associated with the portion of the API source knowledge graph; and

report information describing the defect, the information comprising at least the portion of the API documentation associated with the portion of the API source knowledge graph.

9 . The non-transitory computer-readable medium of claim 8 , wherein the API source knowledge graph comprises a plurality nodes, each node corresponding to a header of an API source document.

10 . The non-transitory computer-readable medium of claim 9 , wherein a given API source document has a given header and has a reference to a different API source document having a different header.

11 . The non-transitory computer-readable medium of claim 10 , wherein the API source knowledge graph comprises an edge between a first node corresponding to the given header and a second node corresponding to the different header based on the reference to the different API source document.

12 . The non-transitory computer-readable medium of claim 8 , wherein the instructions to determine the portion of the connector knowledge representation associated with the defect comprise instructions to:

determine an API feature that is causing the defect; and

identify a node in the API source knowledge graph corresponding to the API feature.

13 . The non-transitory computer-readable medium of claim 12 , wherein the instructions to identify the portion of the API documentation associated with the portion of the API source knowledge graph comprise instructions to:

determine a set of nodes depending from the node in the API source knowledge graph based on there being one or more directional edges pointing from the node to each node of the set of nodes; and

determine the portion to include the set of nodes.

14 . The non-transitory computer-readable medium of claim 13 , wherein the instructions to report the information describing the defect comprise instructions to:

output one or more links to API documentation that corresponds to each node of the set of nodes.

15 . A system for monitoring connectors deployed in a data collection system, the system comprising:

memory with instructions encoded thereon; and

one or more processors that, when executing the instructions, are caused to perform operations comprising:

deploying a data connector generated from a connector knowledge representation, the connector knowledge representation generated from an API source knowledge graph using a large language model;

receiving information describing a defect reported in connection with the deployed data connector;

determining a portion of the connector knowledge representation associated with the defect;

mapping the portion of the connector knowledge representation to a portion of the API source knowledge graph;

identifying a portion of an API documentation associated with the portion of the API source knowledge graph; and

reporting information describing the defect, the information comprising at least the portion of the API documentation associated with the portion of the API source knowledge graph.

16 . The system of claim 15 , wherein the API source knowledge graph comprises a plurality nodes, each node corresponding to a header of an API source document.

17 . The system of claim 16 , wherein a given API source document has a given header and has a reference to a different API source document having a different header.

18 . The system of claim 17 , wherein the API source knowledge graph comprises an edge between a first node corresponding to the given header and a second node corresponding to the different header based on the reference to the different API source document.

19 . The system of claim 15 , wherein determining the portion of the connector knowledge representation associated with the defect comprises:

determining an API feature that is causing the defect; and

identifying a node in the API source knowledge graph corresponding to the API feature.

20 . The system of claim 19 , wherein identifying the portion of the API documentation associated with the portion of the API source knowledge graph comprises:

determining a set of nodes depending from the node in the API source knowledge graph based on there being one or more directional edges pointing from the node to each node of the set of nodes; and

determining the portion to include the set of nodes.

Assignments (2)

Priority Claims (1)

Continuity (1)

References Cited (22)