IP Library Granted Patent US 12,481,665
Granted Patent B1
US 12,481,665 · App. 18/899,531 · Granted Nov 25, 2025

Configurable interactive components for structured query results

Inventors: Keith Gargano (Culver City, CA); Ryan Barlow Dall (Culver City, CA); Stefan Joseph Wojciechowski (Winnetka, CA); Joshua Anthony Marsh (Culver City, CA); Andrew Jeremy Crews (Santa Monica, CA)
Assignee: Amazon Technologies, Inc.
G06F16/24575G06F16/24526G06F16/248G06F21/6218
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,481,665
App. No.
18/899,531
Granted
Nov 25, 2025
Kind
B1
Abstract

Described are systems and processes providing meaningful responses to natural language (NL) prompts issued by a user and providing the responses to the user with configurable interactive components selected based in part on a role of the user that enables the user to engage with the responses in meaningful ways. The configurable interactive components may enable a user to perform certain tasks with output data in a response, such as play content, edit content, and/or perform actions by implementation of software code added to the configurable interactive components. The actions may include implementing approvals, sending electronic messages, and/or scheduling meetings, among many other actions.

Claims (113)

1 . A system, comprising:

one or more processors; and

a memory storing program instructions that when executed by the one or more processors cause the one or more processors to at least:

receive, from a user and by one or more large language model (LLM), a prompt formed from natural language;

determine an organizational role associated with the user;

create, by the one or more LLM, a structured graph query based on at least one word in the prompt and the organizational role, the structured graph query to extract relevant data from a secure data repository that includes a unified graph schema that maps to the relevant data;

send the structured graph query for fulfillment using an application program interface (API);

execute the API to determine the relevant data included in the data repository and associated with the at least one word in the prompt;

receive, by the API and in response to the structured graph query, the relevant data for inclusion with the prompt and for execution by the one or more LLM;

generate, by the one or more LLM, output data as a response to the prompt based at least in part on the relevant data and the organizational role associated with the user;

select at least one configurable interactive component to provide the output data, the at least one configurable interactive component selected based at least in part on the organizational role of the user and the prompt; and

provide the at least one configurable interactive component to the user to provide the user with the output data that is responsive to the prompt.

2 . The system of claim 1 , wherein:

the at least one configurable interactive component provides a user interface to enable the user to access the output data.

3 . The system of claim 1 , wherein the program instructions that when executed by the one or more processors further cause the one or more processors to at least:

receive a user selection from the at least one configurable interactive component; and

implement the user selection to perform at least one of:

create a draft communication to send to another user;

approve a request; or

store edited data in a data repository.

4 . The system of claim 1 , wherein:

the at least one configurable interactive component includes at least one of a table or chart; and

the program instructions that when executed by the one or more processors further cause the one or more processors to at least:

perform at least one data transformation of the relevant data to create the output data that is formatted for the at least one of the table or chart;

populate the table or chart with the output data; and

provide at least one control accessible using the at least one configurable interactive component to enable user interaction with the relevant data.

5 . The system of claim 1 , wherein:

the at least one configurable interactive component includes a permission approval component; and

the program instructions that when executed by the one or more processors further cause the one or more processors to at least:

receive credentials of the user; and

identify a task to be approved using the permission approval component;

determine that the user has authority to approve the task based at least in part on the credentials; and

perform at least one of:

populate the permission approval component to include the task to be approved; or

execute the permission approval component to approve the task.

6 . A method, comprising:

receiving, from a user and by one or more large language model (LLM), a prompt formed from natural language;

determining an organizational role associated with the user;

creating, by the one or more LLM, a structured graph query based on at least one word in the prompt and the organizational role, the structured graph query to extract relevant data from a secure data repository that includes a unified graph schema that maps to the relevant data;

sending the structured graph query for fulfillment using an application program interface (API);

executing the API to determine the relevant data included in the data repository and associated with the at least one word in the prompt;

receiving, by the API and in response to the structured graph query, the relevant data for inclusion with the prompt and for execution by the one or more LLM;

generating, by the one or more LLM, output data as a response to the prompt based at least in part on the relevant data and the organizational role associated with the user;

selecting at least one configurable interactive component to provide the output data, the at least one configurable interactive component selected based at least in part on the organizational role of the user and the prompt; and

providing the at least one configurable interactive component to the user to provide the user with the output data that is responsive to the prompt.

7 . The method of claim 6 , wherein:

the at least one configurable interactive component provides a user interface to enable the user to access the output data.

8 . The method of claim 6 , further comprising:

receiving a user selection from the at least one configurable interactive component; and

implementing the user selection to perform at least one of:

create a draft communication to send to another user;

approve a request; or

store edited data in a data repository.

9 . The method of claim 6 , wherein:

the at least one configurable interactive component includes at least one of a table or chart; and further comprising:

performing at least one data transformation of the relevant data to create the output data that is formatted for the at least one of the table or chart;

populating the table or chart with the output data; and

providing at least one control accessible using the at least one configurable interactive component to enable user interaction with the relevant data.

10 . The method of claim 6 , wherein:

the at least one configurable interactive component includes a permission approval component; and further comprising:

receiving credentials of the user; and

identifying a task to be approved using the permission approval component;

determining that the user has authority to approve the task based at least in part on the credentials; and

performing at least one of:

populating the permission approval component to include the task to be approved; or

executing the permission approval component to approve the task.

11 . The method of claim 6 , further comprising:

creating an executable operation for inclusion with the at least one configurable interactive component to enable the user to cause performance of the executable operation with at least some of the output data.

12 . The method of claim 6 , further comprising:

receiving an input from the user via the at least one configurable interactive component; and

processing the input to cause an executable action to be performed by the at least one configurable interactive component.

13 . The method of claim 6 , further comprising:

receiving, using the at least one configurable interactive component and from the user, an edit to the output data; and

saving the edit in association with the output data.

14 . The method of claim 6 , further comprising:

receiving credentials associated with the user; and

providing the credentials to the at least one configurable interactive component to enable the at least one configurable interactive component to perform at least one permissioned action on behalf of the user.

15 . The method of claim 6 , further comprising:

receiving a user selection from the at least one configurable interactive component; and

implementing the user selection using at least some of the output data.

16 . A computer-implemented method, comprising:

receiving, from a user and by one or more large language model (LLM), a prompt formed from natural language;

determining an organizational role associated with the user;

creating, by the one or more LLM, a structured graph query based on at least one word in the prompt and the organizational role, the structured graph query to extract relevant data from a secure data repository that includes a unified graph schema that maps to the relevant data;

sending the structured graph query for fulfillment using an application program interface (API);

executing the API to determine the relevant data included in the data repository and associated with the at least one word in the prompt;

receiving, by the API and in response to the structured graph query, the relevant data for inclusion with the prompt and for execution by the one or more LLM;

generating, by the one or more LLM, output data as a response to the prompt based at least in part on the relevant data and the organizational role associated with the user;

selecting at least one configurable interactive component to provide the output data, the at least one configurable interactive component selected based at least in part on the organizational role of the user and the prompt; and

providing the at least one configurable interactive component to the user to provide the user with the output data that is responsive to the prompt;

receiving a user selection from the at least one configurable interactive component; and

implementing the user selection to perform at least one of:

create a draft communication to send to another user;

approve a request; or

store edited data in a data repository.

17 . The computer-implemented method of claim 16 , wherein:

the at least one configurable interactive component provides a user interface to enable the user to access the output data.

18 . The computer-implemented method of claim 16 , further comprising:

receiving credentials associated with the user; and

providing the credentials to the at least one configurable interactive component to enable the configurable interactive component to perform at least one permissioned action on behalf of the user.

19 . The computer-implemented method of claim 16 , wherein:

the at least one configurable interactive component includes at least one of a table or chart; and further comprising:

performing at least one data transformation of the relevant data to create the output data that is formatted for the at least one of the table or chart;

populating the table or chart with the output data; and

providing at least one control accessible using the at least one configurable interactive component to enable user interaction with the relevant data.

20 . The computer-implemented method of claim 16 , wherein:

the at least one configurable interactive component includes a permission approval component; and further comprising:

receiving credentials of the user; and

identifying a task to be approved using the permission approval component;

determining that the user has authority to approve the task based at least in part on the credentials; and

performing at least one of:

populating the permission approval component to include the task to be approved; or

executing the permission approval component to approve the task.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 31, 2025
From: GARGANO, KEITH; DALL, RYAN BARLOW; WOJCIECHOWSKI, STEFAN JOSEPH; MARSH, JOSHUA ANTHONY; CREWS, ANDREW JEREMY
To: AMAZON TECHNOLOGIES, INC.
Reel/Frame 071896/0449 →
References Cited (25)
US 9172621B1 · Dippenaar · 2015 [cited by applicant]
US 9268841B2 · Mattox · 2016 [cited by examiner]
US 9582534B1 · Hoover · 2017 [cited by examiner]
US 11763016B2 · Blass · 2023 [cited by examiner]
US 12326869B1 · Alperin · 2025 [cited by examiner]
US 20120253984A1 · Westphal · 2012 [cited by examiner]
US 20190196890A1 · Bucchi et al. · 2019 [cited by applicant]
US 20200004980A1 · Blass · 2020 [cited by examiner]
US 20210144066A1 · Marinucci et al. · 2021 [cited by applicant]
US 20210149851A1 · Belezko et al. · 2021 [cited by applicant]
US 20230045347A1 · Grady · 2023 [cited by applicant]
US 20230289339A1 · Sagar · 2023 [cited by applicant]
US 20230359667A1 · Zionts et al. · 2023 [cited by applicant]
US 20240134857A1 · Gladwin · 2024 [cited by examiner]
US 20240281472A1 · LaRhette · 2024 [cited by examiner]
US 20250139106A1 · Rao Karikurve · 2025 [cited by examiner]
US 20250148289A1 · Mukherjee · 2025 [cited by examiner]
US 20250148308A1 · Vinay · 2025 [cited by examiner]
US 20250217418A1 · Bathwal · 2025 [cited by examiner]
WO 2016138566A1 · 2016 [cited by applicant]
NEO4J, Inc., “Neo4j for GenAI—Ground LLMs with Knowledge Graphs,” www.neo4j.com, Neo4j, Inc. Accessed Sep. 11, 2024. URL: https://neo4j.com/GENERATIVEAI/. [cited by applicant]
NEO4J, Inc., “NeoConverse—Graph Database Search with Natural Language,” www.neo4j.com, Neo4j, Inc. Accessed Sep. 11, 2024, URL: https://neo4j.com/labs/genai-ecosystem/neoconverse/. [cited by applicant]
Ong, Ryan, “Claude Artifacts 101: Types, Use Cases, Sharing, and Remixing,” www.datacamp.com, DataCamp, Inc., Jul. 2024, URL: https://www.datacamp.com/blog/claude-artifacts-introduction, 16 pages. [cited by applicant]
U.S. Appl. No. 17/892,813, filed Aug. 22, 2022, entitled “Unified Graph of Disparate Data Domains,” 57 pages. [cited by applicant]
Vesoft Inc., “NebulaGraph Launches Industry-First Graph RAG: Retrieval-Augmented Generation with LLM Based on Knowledge Graphs,” Sep. 6, 2023, www.nebula-graph.io, NebulaGraph, vesoft Inc., Accessed Sep. 11, 2024, URL: … [cited by applicant]
Cited By (2)
US 12,639,518 US 12,730,903