Granted Patent B2

US 12,498,908 · App. 18/375,371 · Granted Dec 16, 2025

Inventors: Nate Gillman (New York, NY); Nadia Lahlaf (Billerica, MA); Aperahama Parangi (Boston, MA); Jonathon Reilly (Cambridge, MA); Nathan Wies (Brookline, MA)

Assignee: Akkio Inc.

G06F8/35G06N20/00

Abstract

Claims (20)

1 . A method for automatically generating, and executing in real time, computer code using a natural language description of a data manipulation to be performed on a data set, the method comprising:

receiving, by a machine learning engine executing on a computing device, a user-specified data set and a user-specified task;

analyzing, by the machine learning engine, at least one characteristic of the user-specified data set and at least one characteristic of the user-specified task;

generating, by the machine learning engine, at least one machine learning model for processing the user-specified data set, wherein generating further comprises generating the at least one machine learning model based upon the at least one characteristic of the user-specified data set and at least one characteristic of the user-specified task and wherein generating further comprises executing a neural architecture search to progressively build an ensemble of a plurality of machine learning models;

directing, by the machine learning engine, the at least one machine learning model to generate a first output by processing the user-specified data set;

receiving, by the machine learning engine, a natural language description of a second user-requested data transformation task for execution with a subset of the first output;

directing, by the machine learning engine, a large language model to identify an archetype of the user-requested data transformation task;

applying, by the large language model, the user-requested data transformation task to the subset of the first output using the archetype to generate a second output, wherein applying further comprises generating, by a data set preparation engine, executable code for performing the user-requested data transformation task;

executing, by the computing device, the generated executable code with the subset of the first output; and

displaying, by a user interface engine in communication with the machine learning engine, to a user, a user interface including a display of the second output.

2 . The method of claim 1 , wherein receiving the natural language description of the user-requested data transformation task further comprises receiving a prompt to generate executable computer code.

3 . The method of claim 1 further comprising generating, by the large language model, executable computer code in a computer programming language specified in the natural language description of the user-requested data transformation task.

4 . The method of claim 3 wherein applying further comprises executing the generated executable computer code.

5 . A method for automatically generating, and executing in real time, computer code using a natural language description of a data manipulation to be performed on a data set, the method comprising:

receiving, by a machine learning engine executing on a computing device, a user-specified data set and a natural language description of a user-requested data transformation task for execution with a subset of the user-specified data set;

directing, by the machine learning engine, a large language model to generate at least one candidate executable computer code for performing the user-requested data transformation task;

generating, by the large language model in communication with a data set preparation engine, the at least one candidate executable computer code;

performing, by the machine learning engine, at least one validation check on the at least one candidate executable computer code;

executing, by the machine learning engine, the at least one candidate executable computer code to generate a transformation result; and

modifying, by a user interface engine in communication with the machine learning engine, a user interface to display to a user a sample of the transformation result.

Assignments (2)

Continuity (3)

References Cited (64)