Systems and methods for automating generation of information technology project estimates using artificial intelligence/machine learning techniques

1 . A device, comprising:

a processing system including a processor; and

a memory that stores executable instructions that, when executed by the processing system, facilitate performance of operations, the operations comprising:

receiving a user query describing a target information technology (IT) project and a request for a desired action, wherein the desired action includes obtaining an estimate of the target IT project;

via a presentation layer that implements a chatbot interface and includes predetermined logic, prompting a user for additional information to collect data necessary for the target IT project when the user query does not include the data necessary for the target IT project;

maintaining a knowledge repository that stores a data set relevant to IT projects, wherein the data set relevant to IT projects includes at least system information, requirement documentation, financial information, and historical information;

training an artificial intelligence/machine learning (AI/ML) model with a training data set which includes at least a subset of the data set relevant to IT projects;

identifying and extracting, by a large language model of the trained AI/ML model, pertinent information from the user query, wherein the pertinent information identifies an application, a field, a database, and a component associated with the target IT project;

constructing, using the trained AI/ML model, the estimate of the target IT project by using the data set relevant to IT projects stored in the knowledge repository, wherein the constructing the estimate of the target IT project further comprises:

determining, using the pertinent information and accessing the system information stored in the knowledge repository, one or more applications to be impacted by the target IT project;

performing, in an automated sequence following the determining the one or more applications to be impacted by:

scanning code segments, using the system information comprising application codebases and database schema information, that reference one or more fields to be impacted for the one or more applications to be impacted;

determining, based on results of the scanning and the requirement documentation, (i) one or more code modules, database schema objects, API/interface definitions, or user interface elements that reference the one or more fields, and (ii) one or more modifications to the one or more code modules, database schema objects, API/interface definitions, or user interface elements to implement the target IT project in connection with the one or more applications to be impacted; and

determining, based on interface documents of the requirement documentation, where information associated with the one or more fields is sent and which other applications are impacted;

providing information from the automated sequence as a feedback in the determining of the one or more applications to be impacted by the target IT project; and

generating the estimate of the target IT project by performing, in an automated sequence; finding relevant previous estimates, with respect to the determined one or more applications to be impacted, by searching a historical IT project estimate database and a historical actual cost database; determining resources and labor spent in the relevant previous estimates; and computing the estimate of the target IT project by using cost rates and vendor rates stored in a financial information database; and

returning the generated estimate of the target IT project as a response to the user query, wherein the generated estimate includes an application list with impacts identifying areas expected to be changed, wherein the impacts identifying areas expected to be changed comprise one or more of modules, interfaces, database, and user interfaces, and the application list further includes applications delivery use cases and a cost estimate including a cost range.

2 . The device of claim 1 , wherein the system information comprises information relating to database schema, client applications, application codes, application programming interfaces (APIs), user interfaces or a combination thereof; and

wherein the generating the estimate of the target IT project further comprises determining one or more applications impacted in the target IT project based on the system information.

3 . The device of claim 1 , wherein:

the requirement documentation further comprises user stories, interface documents, design documents, interface documents, or a combination thereof; and

the historical information further comprises historical project estimates and historical project actuals.

4 . The device of claim 3 , wherein the generating the estimate of the target IT project further comprises:

identifying prior IT projects having a first predetermined similarity threshold to information included in the requirement documentation and reviewing tasks having performed in the identified prior IT projects; and

identifying and reviewing prior estimates of the IT projects having a second predetermined similarity threshold to information included in the historical information.

5 . The device of claim 1 , wherein the financial information comprises cost rates and vendor rates; and

wherein the generating the estimate of the target IT project further comprises calculating the estimate of the target IT project based on information included in the financial information.

6 . The device of claim 1 , wherein the user query is in a natural language form and the AI/ML model comprises a large language model.

7 . The device of claim 1 , wherein the receiving the user query further comprises receiving the user query via a chatbot interface such that a prompt to the AI/ML model is automatically generated via the chatbot interface.

8 . A non-transitory machine-readable medium, comprising executable instructions that, when executed by a processing system of a user equipment including a processor, facilitate performance of operations, the operations comprising:

receiving a user query describing a target information technology (IT) project and a request for a desired action, wherein the user query contains a plurality of parameters and the desired action includes an instruction to obtain an estimate of the target IT project;

via a presentation layer that implements a chatbot interface and includes predetermined logic, prompting a user for additional information to collect data necessary for the target IT project when the user query does not include the data necessary for the target IT project;

maintaining a knowledge repository that stores a data set relevant to IT projects, wherein the data set relevant to IT projects contains information matching with the plurality of parameters, the information including at least system information, requirement documentation, financial information, and historical inform;

identifying and extracting, by a large language model, pertinent information from the user query, wherein the pertinent information identifies an application, a field, a database, and a component associated with the target IT project;

in response to the user query, constructing, using the large language model, the estimate of the target IT project by accessing the data set relevant to IT projects stored in the knowledge repository, wherein the constructing the estimate of the target IT project comprises:

determining, using the pertinent information and accessing the system information stored in the knowledge repository, one or more applications to be impacted by the target IT project;

performing, in an automated sequence following the determining the one or more applications to be impacted by:

scanning code segments, using the system information comprising application codebases and database schema information, that reference one or more fields to be impacted and one or more other fields likely to be impacted;

determining, based on results of the scanning and the requirement documentation, one or more code modules, database schema objects, API/interface definitions, or user interface elements that reference the one or more fields and one or more modifications to implement the target IT project;

determining, based on interface documents of the requirement documentation, applications to which information associated with the one or more fields is sent; and

providing the determining the applications to be impacted; and

generating the estimate of the target IT project by performing, in an automated sequence; finding relevant previous estimates, with respect to the determined one or more applications to be impacted, by searching a historical IT project estimate database and a historical actual cost database; determining resources and labor spent in the relevant previous estimates; and computing the estimate of the target IT project by using cost rates and vendor rates stored in a financial information database; and

returning the generated estimate of the target IT project as a response to the user query, wherein the estimate includes an application list with impacts identifying areas expected to be changed, wherein the impacts identifying areas expected to be changed comprise one or more of modules, interfaces, database, and user interfaces, and the application list further includes a cost estimate including a cost range.

9 . The non-transitory machine-readable medium of claim 8 , wherein the generating the estimate of the target IT project further comprise decoding the received user query in a natural language form, wherein the plurality of parameters includes applications to be impacted, a prior IT project relevant to the target IT project, a system to be impacted, a type of task or a combination thereof.

10 . The non-transitory machine-readable medium of claim 8 , wherein the accessing the knowledge repository further comprise searching the knowledge repository and retrieving the information matching with the plurality of parameters from the knowledge repository.

11 . The non-transitory machine-readable medium of claim 8 , wherein the data set relevant to IT projects are categorized into at least system information, requirement documentation, financial information, and historical information; and

wherein the accessing the knowledge repository further comprise searching into, and retrieving the information matching with the plurality of parameters from, one or more of system information, requirement documentation, financial information, and historical information.

12 . The non-transitory machine-readable medium of claim 8 , wherein the operations further comprise automating generation of a prompt or instructions to be provided to the large language model as input by guiding the user query to provide the plurality of parameters.

13 . The non-transitory machine-readable medium of claim 8 , wherein the operations further comprise receiving a periodically ingesting, at least system information, requirement documentation, financial information, historical information or a combination thereof, into the knowledge repository.

14 . A method, comprising:

receiving, by a processing system including a processor, a user query describing a target information technology (IT) project and a request for a desired action, wherein the user query contains a plurality of parameters and the desired action includes a request for an estimate of the target IT project;

via a presentation layer that implements a chatbot interface and includes predetermined logic, prompting a user for additional information to collect data necessary for the target IT project when the user query does not include the data necessary for the target IT project;

maintaining, by the processing system, a knowledge repository that stores a data set relevant to IT projects, wherein the data set relevant to IT projects contains information matching with the plurality of parameters, the information including at least system information, requirement documentation, financial information, and historical information;

identifying and extracting, by a large language model of an artificial intelligence/machine learning (AI/ML) model, pertinent information from the user query, wherein the pertinent information identifies an application, a field, a database, and a component associated with the target IT project;

in response to the user query, constructing, by the processing system, using the AI/ML model, the estimate of the target IT project by accessing the data set relevant to IT projects stored in the knowledge repository, wherein constructing the estimate of the target IT project comprises:

determining, using the pertinent information and accessing the system information stored in the knowledge repository, one or more applications to be impacted by the target IT project;

performing, in an automated sequence following the determining the one or more applications to be impacted by:

scanning code segments, using the system information comprising application codebases and database schema information, that reference one or more fields to be impacted and one or more other fields likely to be impacted;

determining, based on results of the scanning and the requirement documentation, one or more code modules, database schema objects, API/interface definitions, or user interface elements that reference the one or more fields and one or more modifications to implement the target IT project;

determining, based on interface documents of the requirement documentation, applications to which information associated with the one or more fields is sent; and

providing the determining the applications to be impacted; and

generating the estimate of the target IT project by performing, in an automated sequence: finding relevant previous estimates, with respect to the determined one or more applications to be impacted, by searching a historical IT project estimate database and a historical actual cost database; determining resources and labor spent in the relevant previous estimates; and computing the estimate of the target IT project by using cost rates and vendor rates stored in a financial information database; and

providing, by the processing system, the generated estimate of the target IT project as a response to the user query, wherein the estimate includes an application list with impacts identifying areas expected to be changed, wherein the impacts identifying areas expected to be changed comprise one or more of modules, interfaces, database, and user interfaces, and the application list further includes a cost estimate including a cost range.

15 . The method of claim 14 , wherein the generating the estimate of the target IT project further comprise:

encoding and decoding, by the processing system, the received user query in a natural language form to be provided to the AI/ML model as input; and

in response to the user query in the natural language form, generating, by the processing system, the estimate of the target IT project as output, using a large language model as the AI/ML model.

16 . The method of claim 14 , wherein the accessing the knowledge repository further comprise searching the knowledge repository and retrieving the information matching with the plurality of parameters from the knowledge repository.

17 . The method of claim 14 , wherein the data set relevant to IT projects are categorized into at least system information, requirement documentation, financial information, and historical information; and

wherein the accessing the knowledge repository further comprise searching into and retrieving the information matching with the plurality of parameters from one or more of system information, requirement documentation, financial information, and historical information.

18 . The method of claim 14 , further comprising automating, by the processing system, generation of a prompt or instructions to be provided to the AI/ML model as input by guiding the user query to provide the plurality of parameters via a chatbot interface.

19 . The method of claim 14 , further comprising receiving, by the processing system, periodically ingesting, at least system information, requirement documentation, financial information, and historical information, into the knowledge repository.

20 . The method of claim 14 , further comprising training, by the processing system, the AI/ML model using system information, requirement documentation, financial information, and historical information,

wherein the system information comprises information relating to database schema, client applications, application codes, application programming interfaces (APIs), user interfaces or a combination thereof;

the requirement documentation further comprises user stories, interface documents, design documents, interface documents, or a combination thereof; and

the historical information further comprises historical project estimates and historical project actuals.