Description:TECHNICAL FIELD
[001] This disclosure relates generally to adaptive learning of Artificial Intelligence (AI) agents, more particularly to method and system for dynamic skilling of the AI agents for providing solutions for Information Technology (IT) operations.

BACKGROUND
[002] In the domain of Information Technology (IT), in general, organizations use pre-trained AI agents to automate repetitive tasks, increase efficiency, and to resolve IT issues of the IT operations. By automating routine and time-consuming tasks through AI agents, organizations save time and resources. Through these human resources of the organization can focus on working on complex IT tasks that require human involvement.
[003] In the present state of the art, the AI agents are trained on large datasets pertaining to IT operations to solve problems that arise during the IT operations. While the AI agents can provide solutions for problems that are known using their pre-existing knowledge. As the IT operations domain is constantly evolving. Sometimes the AI agents might receive new problems from the users that are outside their pre-existing knowledge. The AI agents may struggle to provide accurate solutions to the new problems that are outside their initial training data. To address these issues the AI agents should dynamically learn and update their training knowledge to be able to provide solutions for the new problems.
[004] Therefore, there is a need to provide systems and methods for dynamic skilling of the AI agents.

SUMMARY
[005] In one embodiment, a method for dynamic skilling of an Artificial Intelligence (AI) agent is disclosed. In one example, the method includes receiving a first request from a user. The first request indicates a first technical obstacle that requires diagnostic analysis and troubleshooting. The method further includes determining, by the AI agent, the necessitation of a first skill, and accessing one or more data sources to obtain a primary dataset associated with the first skill. The method also includes validating the first skill, prior to application, based on one or more operational parameters. The application of the first skill is associated with the resolution of the first technical obstacle. The method further includes receiving a second request from the user. The second request is indicative of a second technical obstacle requiring diagnostic analysis and troubleshooting. The method further includes determining the necessitation of a second skill, pruning the primary dataset prior to accessing the one or more data sources to obtain a secondary dataset pertaining to the second skill, compiling the pruned primary dataset and the secondary dataset via an integration engine, and validating and applying the second skill based on the one or more operational parameters to generate a resolution for the second technical obstacle.
[006] In one embodiment, a system for dynamic skilling of an AI agent is disclosed. In one example, the system includes a processor and a memory communicatively coupled to the processor. The memory stores processor-executable instructions, which, when executed by the processor, cause the processor to perform operations including receiving a first request from a user indicative of a first technical obstacle requiring diagnostic analysis and troubleshooting. The instructions further cause the processor to determine the necessitation of a first skill, access one or more data sources to obtain a primary dataset for the first skill, and validate the first skill prior to its application based on one or more operational parameters. The instructions also cause the processor to receive a second request from the user for a second technical obstacle, determine the necessitation of a second skill, prune the primary dataset, access the secondary dataset from one or more data sources, and compile the datasets via an integration engine. The compiled data is then validated and used for applying the second skill to resolve the second technical obstacle.
[007] In one embodiment, a non-transitory computer-readable medium storing computer-executable instructions for dynamic skilling of an AI agent is disclosed. In one example, the stored instructions, when executed by a processor, cause the processor to receive a first user request indicative of a first technical obstacle requiring diagnostic analysis and troubleshooting. The instructions further cause the processor to determine the necessitation of a first skill, access one or more data sources to retrieve a primary dataset for the first skill, and validate the skill prior to its application based on operational parameters. The instructions also cause the processor to receive a second request indicating a second technical obstacle, determine the necessitation of a second skill, prune the primary dataset, retrieve a secondary dataset from the one or more data sources, compile the datasets via an integration engine, and validate and apply the second skill to resolve the second technical obstacle.
[008] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.

BRIEF DESCRIPTION OF THE DRAWINGS
[009] The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the disclosed principles.
[010] FIG. 1 illustrates an exemplary ecosystem of a plurality of users interacting with an AI assistance system, in accordance with an example embodiment.
[011] FIG. 2 illustrates an AI agent execution system, in accordance with an example embodiment.
[012] FIG. 3 illustrates an AI agent management system, in accordance with an example embodiment.
[013] FIG. 4 illustrates an AI agent interface system, in accordance with an example embodiment.
[014] FIG. 5 is a block diagram of an exemplary computer system for implementing embodiments consistent with the present disclosure.

DETAILED DESCRIPTION
[015] Exemplary embodiments are described with reference to the accompanying drawings. Wherever convenient, the same reference numbers are used throughout the drawings to refer to the same or like parts. While examples and features of disclosed principles are described herein, modifications, adaptations, and other implementations are possible without departing from the spirit and scope of the disclosed embodiments. It is intended that the following detailed description be considered as exemplary only, with the true scope and spirit being indicated by the following claims.
[016] Referring to FIG. 1, an exemplary multi-AI agent troubleshooting ecosystem 100 is illustrated, in accordance with an example embodiment of the present disclosure. The ecosystem 100 illustrates a plurality of users 104 interacting with an AI assistance system 102. The AI assistance system 102 includes a user interface 108a, 108b, 108c. The user interface 108a, 108b, 108c may capture user inputs, translates them into structured queries, and facilitates seamless communication between human users and the AI agents. In some embodiments, each of the plurality of users 104 may have an individual user interface 108a, 108b, 108c by means of a personal computer, a mobile device or a tablet device. In some embodiments, the user interface 108a, 108b, 108c could be same for every user. The AI assistance system 102 further includes an AI agent platform 106, a plurality of AI agents 110, one or more data sources 112, a database server 114, an AI dashboard 116. The plurality of users 104a, 104b, 104c may include but not limited to an information technology service management (ITSM) engineer, network operations (NetOps) engineer, cloud operations (CloudOps) engineer, database engineer, software developer, DevOps engineer, site reliability engineer (SRE), backend developer, frontend developer, full stack developer, quality assurance (QA) engineer, software architect, mobile application developer, data engineer, automation engineer, systems programmer, or any other technical professional involved in software development, support, process operations, Information Technology (IT) operations, deployment, testing, or maintenance. The AI assistance system 102 provides assistance to the to the plurality of users 104 in resolving technical obstacles, automating repetitive tasks pertaining to the IT operations of an organization.
[017] In some embodiments, one or more users of the plurality of users 104 may transmit a first request to the AI agent platform 106 by interacting with the user interface 108. The first request may be related to resolving a technical obstacle, automating a repetitive task, increasing efficiency of a task. The technical obstacle may be related to, but not limited to, software development, software support, process operations, and IT operations of the organization. As used herein, the "technical obstacle" may refer to any malfunction, anomaly, error condition, or performance degradation within an information technology (IT) infrastructure that adversely impacts the operation, availability, integrity, or efficiency of a system, component, service, or network. The technical obstacle may arise from hardware faults, software bugs, configuration issues, compatibility conflicts, security breaches, or any other technical cause that necessitates diagnostic evaluation and corrective action.
[018] In some embodiments, the first request may pertain to software development operations such as requirements planning and design, coding, testing, DevOps, and legacy modernization. In some embodiments, the first request may pertain to support operations, such as ticket intelligence, runbook automation, and root cause analysis. In some embodiments, the first request may pertain to process operations such as agent productivity, order-to-cash and marketing operations. In some embodiments, the first request may pertain to IT operations such as hybrid cloud operations, enterprise monitoring and service desk.
[019] In some embodiments, the plurality of AI agents 110a, 110b, 110c may be configured to handle a specific domain of tasks and requests. In some embodiments, the plurality of AI agents 110a, 110b, 110c may include, but are not limited to, an ITSM (Information Technology Service Management) AI agent, a database AI agent, and a NetOps (Network Operations) AI agent, each specialized in addressing domain-specific requests and technical obstacles. In some embodiments, each of the plurality of AI agents 110 are pre-trained on large datasets, each dataset pertaining to the domain-specific knowledge of each of the IT operations.
[020] In some embodiments, the AI agent platform 106 includes the plurality of AI agents 110. The AI agent platform 106 upon receiving the first request for the AI agent 110a, the AI agent 110a generates a response to the first request. The response may include but not limited to a technical resolution, a workflow automation procedure or an optimization recommendation based on the first request. In an example, The AI agent 110a may function as a CloudOps AI agent configured to receive and respond to requests from cloud operations engineer or cloud operations team or any requests pertaining to cloud infrastructure. In some embodiments, the plurality of AI agents 110 may operate within the AI agent platform 106, the working of the AI agent platform 106 will be described in detail with reference to FIG. 2, in subsequent embodiments.
[021] In some embodiments, the AI agent platform 106 may receive a second request for the AI agent 110b outside the scope of the training knowledge of the AI agent 110b. Upon identifying by the AI agent platform 106, that the second request is outside the scope of the training knowledge of the AI agent 110b. The AI agent 110b learns new skills required to provide response for the second request from the one or more data sources 112a, 112b, 112c. The one or more data sources 112a, 112b, 112c may include one or more large language models (LLMs). The one or more LLMs may include, but not limited to Open AI Generative Pre-trained Transformer (GPT) 4, Open AI GPT 4o, Open AI GPT 4o Mini, Open AI GPT 4 Turbo, Open AI GPT 3.5 Turbo, Open AI GPT 3.5 Turbo 1106, Ollama3.1, Ollama3.2.
[022] In some embodiments, the AI agent 110b after acquiring the new skills from the one or more data sources 112a, 112b, 112c. The AI agent 110b stores the data pertaining to the new skills to provide response for the second request in the database server 114. The database server 114 is configured to manage and store data acquired from the one or more LLMs. Upon acquiring the new skills from the data in the database server 114. The AI agent 110b provides accurate response to the second request.
[023] In some embodiments, the AI dashboard 112 may be configured to show the training status of the plurality of AI agents 110 to the plurality of users 104.
[024] In an example, the AI agent 110b is a database AI agent. A database engineer 104b submits a request using the user interface 108b. The request is regarding a technical obstacle related to database optimization. Upon receiving this request, the AI agent platform 106 determines that the request falls outside the scope of the pre-trained knowledge of the database AI agent 110b. Upon determining, the AI agent 110b initiates a skill acquisition process by accessing one or more data sources 112a, 112b, 112c, which include one or more large language models (LLMs) such as OpenAI GPT-4 and Ollama3.1. During the skill acquisition, the AI dashboard 116 dynamically displays the training status of the Database AI agent 110b, indicating percentage of new skills acquired such as 50%, 60%, and so on. Upon acquiring 100% of the new skill, the data related to the new skills acquired from the one or more LLMs is stored in the database server 114. The Database AI agent 110b uses the new skills to provide a resolution to the request submitted by the database engineer 104b.
[025] Referring to FIG.2, an AI agent execution system 200 is illustrated, in accordance with an example embodiment of the present disclosure. The AI agent execution system 200 comprises an AI agent platform 202, a memory 204, a processor 206, one or more data sources 207. The AI agent platform 202 further comprises an AI agent studio 208, within which an AI agent 210 is developed, configured, and trained. The AI agent 210 is operable within the AI agent studio 208 to perform designated tasks. In certain embodiments, the AI agent platform 202 may correspond to, or be functionally similar to, the AI agent platform 106 as described with reference to FIG. 1.. The AI agent 210 corresponds to the AI agent 110 disclosed in description of FIG.1. The processor 206 is communicably coupled to the AI agent 210. The memory 204 is communicably coupled to the processor 206. The memory 204 stores processor instructions. The processor instructions stored in the memory 204 are executed by the processor to execute one or more functions by the AI agent 210. In some embodiments, the AI agent 210 receives a first request from the user 212 for a resolution for a first technical obstacle of the one or more technical obstacles. As used herein the "resolution" refers to the corrective action, output, or remedial response generated by an AI agent in response to a detected or reported technical obstacle within an information technology (IT) infrastructure. The resolution may include, but is not limited to, executing predefined procedures, generating configuration changes, providing diagnostic insights, initiating repair protocols, or delivering guided instructions aimed at restoring normal system functionality, improving performance, or preventing recurrence of the identified technical obstacle. The one or more technical obstacles may include but not limited to functional bugs, logical bugs, security bugs, performance bugs, syntax errors, usability bugs, compatibility bugs, unit level bugs, workflow bugs, system-level integration bugs, data bugs, content bugs, code bugs, and concurrency bugs. The AI agent studio further comprises a skill identification module 214, an interpretation module 215, a data source accessor 216, a skill validation module 218, an integration engine 220, and a data optimization module 222.
[026] In some embodiments, the AI agent 210 sends the first request to the skill identification module 214. The skill identification module 214 identifies the skills required to provide the resolution for the first technical obstacle. The skill identification module 214 checks whether the AI agent 210 has the skills required to provide the resolution for the first technical obstacle. If the skill identification module 214 determines that the AI agent 210 does not possess the skills required to provide the resolution for the first technical obstacle. In some embodiments, the first skill correspond to data representation beyond a coverage of a pre-trained data of the AI agent 210. The skill identification module 214 sends the first request to a data source accessor 216. The data source accessor 216 accesses the one or more data sources 207 through application programming interface (API) integration, to retrieve a primary dataset pertaining to the first skill. In some embodiments, the primary dataset includes criteria-based data from one or more LLMs pertaining to the first skill. In some embodiments, accessing one or more data sources comprises querying the one or more LLMs to retrieve the primary dataset. The querying comprises generating one or more structured requests based on functional requirements associated with obtaining the first skill. In some embodiments, accessing the one or more data sources further comprises parsing by the interpretation module 215 one or more structured requests to extract skill-based attributes for retrieving the primary dataset. In an example, the AI agent may require a first skill pertaining to a technical obstacle of a network engineer. In this example, the interpretation module 215 extracts network metrics and generates one or more structured requests to the one or more LLMs based on the network metrics. In some embodiments, the one or more data sources 207 may include one or more large language models (LLMs). The one or more LLMs may include but not limited to Open AI Generative Pre-trained Transformer (GPT) 4, Open AI GPT 4o, Open AI GPT 4o Mini, Open AI GPT 4 Turbo, Open AI GPT 3.5 Turbo, Open AI GPT 3.5 Turbo 1106, Ollama3.1, Ollama3.2.
[027] In some embodiments, the one or more data sources 207 may be selected from among one or more large language models (LLMs). The AI agent 210 acquires a first skill based on a corresponding primary dataset. A skill validation module 218 is configured to validate the acquired first skill in accordance with one or more operational parameters. In certain embodiments, the validation process includes evaluating the performance of the AI agent 210 against predefined performance metrics to determine eligibility for skill certification. By way of example, the AI agent 210 may acquire the first skill and achieve a performance score of 85%, whereas the predefined threshold for validation is 90%. In such a case, the first skill is deemed not validated. Conversely, if the performance score is determined to be 95%, the first skill is considered successfully validated. In some embodiments, the process of validating the first skill further includes pruning the primary dataset using an iterative moderation technique. The iterative moderation technique may include, but is not limited to, automated review cycles and feedback loops. The operational parameters used in skill validation may include, without limitation, AI agent performance metrics, resolution response time, resolution accuracy, and integration efficiency of the associated integration engine. Following successful validation by the skill validation module 218, the AI agent 210 applies the validated first skill to resolve a corresponding technical obstacle.
[028] As used herein, "performance metrics" may refer to a set of quantifiable parameters utilized to assess, validate, and optimize the operational effectiveness of an AI agent within an AI-driven troubleshooting environment. In some embodiments, the performance metrics may be dynamically generated and contextually adaptive based on the nature of technical obstacles, system architecture, and historical resolution data. The performance metrics may include, but are not limited to: resolution accuracy, defined as the degree to which the AI agent's output aligns with a verified correct solution; response time, measured as the duration between obstacle detection and resolution output; resolution success rate across varying infrastructure contexts; integration efficiency, which quantifies the AI agent’s ability to interface with external systems or modules; and cognitive load optimization, which measures the reduction in human intervention achieved by autonomous AI operations.
[029] In certain embodiments, the performance metrics are leveraged not only for skill validation but also for iterative refinement of AI agent behavior, contributing to a continuous learning loop that distinguishes the ecosystem 100 from rule-based automation frameworks.
[030] In some embodiments, the AI agent 210 receives a second request from the user 212 for a second resolution for a second technical obstacle of the one or more technical obstacles. In some embodiments, the AI agent 210 transforms the first request and the second request in real-time using one or more data anonymization techniques. The data anonymization techniques may include but not limited to data masking, pseudonymization, generalization, suppression, differential privacy and tokenization. The AI agent 210 sends the second request to the skill identification module 214. The skill identification module 214 identifies the skills required to provide the second resolution for the second technical obstacle. The skill identification module 214 checks whether the AI agent 210 has the skills required to provide the second resolution for the second technical obstacle. If the skill identification module 214 determines that the AI agent 210 does not possess the skills required to provide the second resolution for the second technical obstacle. In some embodiments, the second skill correspond to data representation beyond the coverage of the pre-trained data of the AI agent 210. The skill identification module 214 sends information to the data optimization module 222 implying necessitation of a second skill. The data optimization module 222 is communicably coupled to the memory 204, to optimize the memory usage the data optimization module 222, before acquiring a second skill by the AI agent 210 to provide the resolution for the second technical obstacle, the data optimization module 222 prunes the primary dataset to generate a pruned primary dataset. As used herein "pruning" may refer to the process of selectively filtering, removing, or de-prioritizing redundant, irrelevant, outdated, or low-utility data entries from a primary dataset in order to generate a refined subset—referred to as a pruned primary dataset. The pruning operation is performed by the data optimization module 222 and is configured to enhance memory efficiency, reduce computational overhead, and improve the relevance and precision of skill acquisition by the AI agent 210. In some embodiments, pruning may involve heuristic analysis, pattern recognition, contextual relevance scoring, or rule-based exclusion criteria. The pruned primary dataset serves as an optimized input for enabling the AI agent 210 to acquire a second skill, thereby facilitating accurate and resource-efficient resolution of a second technical obstacle.
[031] Then the skill identification module 214 sends the second request to the data source accessor 216 for retrieving a secondary dataset pertaining to the second skill. In some embodiments, the secondary dataset includes criteria-based data from the one or more LLMs pertaining to the second skill. The data source accessor 216 accesses the one or more data sources 207 to retrieve the secondary dataset pertaining to the second skill. In some embodiments, accessing of the one or more data sources comprises querying one or more LLMs to retrieve the secondary dataset. The querying comprises generating one or more structured requests based on functional requirements associated with obtaining the second skill. In some embodiments, accessing the one or more data sources further comprises parsing by the interpretation module 215 one or more structured requests to extract skill-based attributes for retrieving the secondary dataset. After retrieving the secondary dataset. The integration engine 220 compiles the pruned primary dataset and the secondary dataset. The AI agent 210 acquires the second skill using the secondary dataset. The skill validation module 218 validates the second skill acquired by the AI agent 210 according to the one or more operational parameters. The AI agent 210 applies the second skill to provide the second resolution for the second technical obstacle after the validation of the second skill by the skill validation module 218.
[032] Referring now to FIG. 3, an AI agent management system 300 is illustrated, in accordance with some embodiments. In some embodiments, the AI agent management system 300 includes an AI agent studio 302, an AI agent runtime engine 304, a value realization framework 306, an AI dashboard 308, and an AI hub 310. In some embodiments, the AI agent studio 302 corresponds to the agent studio 208 disclosed in description of FIG.2. The AI agent studio 302 serves as a primary interface for the plurality of AI agents’ lifecycle management. The lifecycle of the plurality of AI agents includes development, training, testing, and deployment using a graphical environment. The AI agent studio 302 may utilize React library for the graphical environment to provide a robust, user-friendly development environment for scripting and modeling of the plurality of AI agents. The AI agent studio 302 is configured to interact with various middleware to enhance connectivity and functionality across different layers. In an example, the middleware may be used to connect a development environment and a production environment of the AI agent.
[033] In some embodiments, the AI agent runtime engine 304 is communicatively coupled to the AI agent studio 302. The AI agent runtime engine 304 is configured to execute the plurality of AI agents across one or more operational environments. In an example, the one or more operational environments may include development environment, testing environment, production environment, Windows environment, Linux environment, mac Operating System (OS) environment, iOS, Android, public cloud, private cloud, hybrid cloud, Sandbox environment, restricted access environment, open/public environment. In some embodiments, the AI agent runtime engine 304 is further configured to integrate with third-party Information Technology Service Management (ITSM) platforms, such as ServiceNow, to enable the plurality of AI agents operate across one or more diverse systems. The AI agent runtime engine 304 further configured to utilize Application programming Interfaces (APIs) to connect the ITSM platforms to improve automation capabilities of the plurality of AI agents. The automation capabilities may include enabling the plurality of AI agents to perform actions based on real-time events or requests received from the user in real-time.
[034] In some embodiments, the value realization framework 306 is configured to quantify and analyze the impact of AI interventions and provide stakeholders with measurable outcomes and value assessments. The value realization framework 306 includes an analytics engine. The analytics engine executes advanced mathematical models and analytics to calculate return on investment (ROI) and other key performance indicators (KPIs). The advanced mathematical models may include but not limited to, regression analysis, time series forecasting, decision trees, cluster analysis, Bayesian inference, Monte Carlo simulation, and optimization algorithms. The KPIs may include but not limited to, Incident Management (%), Mean Time to Repair (MTTR), Root Cause Analysis (RCA), Automation Development Speed, Machine Generated RCA (%), Resolution Time for Complex Issues, Efficiency in Automation, Scalability, Error Rate, Cost Efficiency, Automation Coverage, Human Dependency, Security Risk Mitigation, and Tool and API Reusability.
[035] In an example, in traditional systems for incident management, manual intervention is needed more than 50% of the cases. After using the AI agent management system 300 autonomous incident management exceeds 65%. Similarly in other examples, the Mean Time to Repair (MTTR), which traditionally ranged between 4 to 6 hours per incident, is reduced to 1–2 hours due to quicker RCA and decision-making. Root Cause Analysis (RCA), which previously required mostly manual processes taking days, is now machine-assisted with over 95% of RCA tasks automated. Automation development speed, which traditionally involves weeks to months of manual coding, is accelerated to days or hours through automated skill creation. Machine-generated RCA, previously minimal, now results in more than 95% of analysis. Resolution time for complex issues, which could take several days to weeks, is significantly reduced to days or even hours. Efficiency in automation improves from being limited to predefined scenarios to enabling dynamic and adaptive capabilities. Scalability, traditionally requires manual updates and scaling, becomes automatic with real-time adjustments and autonomous skill creation. The error rate, traditionally high due to manual interventions, is lowered through continuous AI-driven improvements. Cost efficiency is achieved by reducing human labor dependency. Automation coverage expands from specific areas to broad, dynamic coverage across systems and tools. Human dependency, which was high in traditional systems, is reduced as the plurality of AI agents become self-trained and certified. Security risk mitigation shifts from slow, manual responses to real-time threat detection and response. Tool and API reusability, traditionally minimal reusability, now AI agent management system 300 reusability increases due to reusable integrations and dynamic tool connections.
[036] In some embodiments, the AI dashboard 308 is operably coupled to each of the components of the AI agent management system 300 and configured to provide a centralized visualization tool to monitor and manage the performance and efficiency of the plurality of AI agents in real-time. The AI dashboard 308 is further configured to provide a set of configurable widgets and scorecards, which may be dynamically updated based on system health in real time. The widgets and scorecards provide insights into one or more operational metrics. In some embodiments, the one or more operational metrics may include but not limited to the KPIs.
[037] In some embodiments, the AI hub 310 is configured to provide version control and distribution management of the plurality of AI agents across the one or more operational environments. In some embodiments, the AI hub 310 utilizes GitOps for version control and deployment of the plurality of AI agents across the one or more operational environments. The AI hub 310 further comprises configurations and settings for the plurality of AI agents. The configurations and settings may be utilized to customize each of the plurality of the AI agents and to update and maintain the latest versions of the plurality of the AI agents.
[038] FIG. 4 illustrates an AI agent interface system 400, the AI agent interface system 400 corresponds to the user interface 108 mentioned in FIG. 1. The AI agent interface system 400 includes a studio interface 402. The studio interface 402 is associated with the AI agent studio 402. The AI agent interface system 400 further includes a runtime interface 404 that is associated with the AI agent runtime engine. 404, a frame interface 406 associated with the value realization framework 406, an AI hub interface 408, and an AI dashboard interface 410. The studio interface 402 is configured to provide an option for uploading one or more documents. The uploaded document(s) are required to include information pertaining to a desired skill set, generic cognitions, and contextualization. The studio interface 402 further comprises a plurality of functional modules, including: an AI Agent Build and Configure Module, a Playground Module, and a Cognition Build and Versioning Module.
[039] In some embodiments, the runtime interface 404 comprises an endpoint setup module. The endpoint setup module is configured to provide access to an integrations library of more than 120 integrations to interact and operate tools and customer investments. In an example, the list of integrations may include but not limited to Information Technology Service Intelligence (ITSI) Integration, Hyper Automation Integration, Robotic Process Automation (RPA) Integration, Customer Relationship Management (CRM) Integration, Enterprise Resource Planning (ERP) Integration, Cloud Infrastructure Integration (e.g., Amazon Web Services (AWS) Integration, Microsoft Azure Integration, Google Cloud Platform (GCP) Integration), and Communication Platform Integration (e.g., Slack Integration, Microsoft Teams Integration, Zoom Integration).
[040] In some embodiments, the AI hub interface 406 comprises an AI agent gallery module. The AI agent gallery module comprises includes a plurality of AI agent profiles, each representing a distinct AI agent that is specifically trained using the AI Agent Studio. Each AI agent profile is associated with a corresponding AI agent trained on domain-specific data sets to address and resolve issues within a particular technological domain. In an example, the AI agent gally module may include but not limited to a first AI agent as a Cloud Engineer AI Agent, a second AI agent as a DevOps Engineer AI Agent, a third AI agent as a CloudOps Engineer AI Agent, a fourth AI agent as an IT Service Management (ITSM) AI Agent, and a fifth AI agent as a Network Operations (NetOps) AI Agent. The user 104 may interact with any of the AI agents from the AI agent gallery module through a chat interface to request a resolution for a domain specific technical obstacle.
[041] In some embodiments, the framework interface 408 comprises a performance evaluation module. The performance evaluation module is configured to extract details from one or more other components of the AI agent interface system 400. In some embodiments, the performance evaluation module is further configured to provide Key Performance Indicator (KPI) and metrics mapping and custom impact mapping based on the extracted data from the other components of the AI agent interface system 400. The performance evaluation module further configured to analyze the KPI and metrics mapping to generate and transmit a performance report to the AI dashboard interface 410. The interface of AI dashboard interface 410 is configured to show insights. The insights may include but not limited to AI financial operations (AI FinOps), AI agent performance metrics and operational gains from the use of the plurality of AI agents.
[042] As will be also appreciated, the above described techniques may take the form of computer or controller implemented processes and apparatuses for practicing those processes. The disclosure can also be embodied in the form of computer program code containing instructions embodied in tangible media, such as floppy diskettes, solid state drives, CD-ROMs, hard drives, or any other computer-readable storage medium, wherein, when the computer program code is loaded into and executed by a computer or controller, the computer becomes an apparatus for practicing the invention. The disclosure may also be embodied in the form of computer program code or signal, for example, whether stored in a storage medium, loaded into and/or executed by a computer or controller, or transmitted over some transmission medium, such as over electrical wiring or cabling, through fiber optics, or via electromagnetic radiation, wherein, when the computer program code is loaded into and executed by a computer, the computer becomes an apparatus for practicing the invention. When implemented on a general-purpose microprocessor, the computer program code segments configure the microprocessor to create specific logic circuits.
[043] The disclosed methods and systems may be implemented on a conventional or a general-purpose computer system, such as a personal computer (PC) or server computer. Referring now to FIG. 5, an exemplary computing system 500 that may be employed to implement processing functionality for various embodiments (e.g., as a SIMD device, client device, server device, one or more processors, or the like) is illustrated. Those skilled in the relevant art will also recognize how to implement the invention using other computer systems or architectures. The computing system 500 may represent, for example, a user device such as a desktop, a laptop, a mobile phone, personal entertainment device, DVR, and so on, or any other type of special or general-purpose computing device as may be desirable or appropriate for a given application or environment. The computing system 500 may include one or more processors, such as a processor 502 that may be implemented using a general or special purpose processing engine such as, for example, a microprocessor, microcontroller or other control logic. In this example, the processor 502 is connected to a bus 504 or other communication medium. In some embodiments, the processor 502 may be an Artificial Intelligence (AI) processor, which may be implemented as a Tensor Processing Unit (TPU), or a graphical processor unit, or a custom programmable solution Field-Programmable Gate Array (FPGA).
[044] The computing system 500 may also include a memory 506 (main memory), for example, Random Access Memory (RAM) or other dynamic memory, for storing information and instructions to be executed by the processor 502. The memory 506 also may be used for storing temporary variables or other intermediate information during execution of instructions to be executed by the processor 502. The computing system 500 may likewise include a read only memory (“ROM”) or other static storage device coupled to bus 504 for storing static information and instructions for the processor 502.
[045] The computing system 500 may also include a storage devices 508, which may include, for example, a media drive 510 and a removable storage interface. The media drive 510 may include a drive or other mechanism to support fixed or removable storage media, such as a hard disk drive, a floppy disk drive, a magnetic tape drive, an SD card port, a USB port, a micro USB, an optical disk drive, a CD or DVD drive (R or RW), or other removable or fixed media drive. A storage media 512 may include, for example, a hard disk, magnetic tape, flash drive, or other fixed or removable medium that is read by and written to by the media drive 510. As these examples illustrate, the storage media 512 may include a computer-readable storage medium having stored therein particular computer software or data.
[046] In alternative embodiments, the storage devices 508 may include other similar instrumentalities for allowing computer programs or other instructions or data to be loaded into the computing system 500. Such instrumentalities may include, for example, a removable storage unit 514 and a storage unit interface 516, such as a program cartridge and cartridge interface, a removable memory (for example, a flash memory or other removable memory module) and memory slot, and other removable storage units and interfaces that allow software and data to be transferred from the removable storage unit 514 to the computing system 500.
[047] The computing system 500 may also include a communications interface 518. The communications interface 518 may be used to allow software and data to be transferred between the computing system 500 and external devices. Examples of the communications interface 518 may include a network interface (such as an Ethernet or other NIC card), a communications port (such as for example, a USB port, a micro USB port), Near field Communication (NFC), etc. Software and data transferred via the communications interface 518 are in the form of signals which may be electronic, electromagnetic, optical, or other signals capable of being received by the communications interface 518. These signals are provided to the communications interface 518 via a channel 520. The channel 520 may carry signals and may be implemented using a wireless medium, wire or cable, fiber optics, or other communications medium. Some examples of the channel 520 may include a phone line, a cellular phone link, an RF link, a Bluetooth link, a network interface, a local or wide area network, and other communications channels.
[048] The computing system 500 may further include Input/Output (I/O) devices 522. Examples may include, but are not limited to a display, keypad, microphone, audio speakers, vibrating motor, LED lights, etc. The I/O devices 522 may receive input from a user and also display an output of the computation performed by the processor 502. In this document, the terms “computer program product” and “computer-readable medium” may be used generally to refer to media such as, for example, the memory 506, the storage devices 508, the removable storage unit 514, or signal(s) on the channel 520. These and other forms of computer-readable media may be involved in providing one or more sequences of one or more instructions to the processor 502 for execution. Such instructions, generally referred to as “computer program code” (which may be grouped in the form of computer programs or other groupings), when executed, enable the computing system 500 to perform features or functions of embodiments of the present invention.
[049] In an embodiment where the elements are implemented using software, the software may be stored in a computer-readable medium and loaded into the computing system 500 using, for example, the removable storage unit 514, the media drive 510 or the communications interface 518. The control logic (in this example, software instructions or computer program code), when executed by the processor 502, causes the processor 502 to perform the functions of the invention as described herein.
[050] As will be appreciated by those skilled in the art, the techniques described in the various embodiments discussed above are not routine, or conventional, or well understood in the art. The techniques discussed above provide for dynamic skilling of an AI agent. The techniques first receive a first request from a user, the first request being indicative of a first technical obstacle requiring diagnostic analysis and troubleshooting. The techniques then determine necessitation of obtaining a first skill. The techniques then access one or more data sources to retrieve a primary dataset pertaining to the first skill. The techniques then validate the first skill prior to application according to one or more operational parameters. The application is associated with resolution of the first technical obstacle. The techniques then receive a second request from the user, the second request being indicative of a second technical obstacle requiring diagnostic analysis and troubleshooting. The techniques then determine necessitation of obtaining a second skill. The techniques then prune the primary dataset prior to accessing the one or more data sources to retrieve secondary dataset pertaining to the second skill. The techniques then compile pruned primary dataset and the secondary dataset for validation and application of the second skill based on the one or more operational parameters. The techniques then generate a resolution for the second technical obstacle.
[051] In light of the above mentioned advantages and the technical advancements provided by the disclosed method and system, the claimed steps as discussed above are not routine, conventional, or well understood in the art, as the claimed steps enable the following solutions to the existing problems in conventional technologies. Further, the claimed steps clearly bring an improvement in the functioning of the device itself as the claimed steps provide a technical solution to a technical problem.
[052] The specification has described method and system for dynamic skilling of an AI agent. The illustrated steps are set out to explain the exemplary embodiments shown, and it should be anticipated that ongoing technological development will change the manner in which particular functions are performed. These examples are presented herein for purposes of illustration, and not limitation. Further, the boundaries of the functional building blocks have been arbitrarily defined herein for the convenience of the description. Alternative boundaries can be defined so long as the specified functions and relationships thereof are appropriately performed. Alternatives (including equivalents, extensions, variations, deviations, etc., of those described herein) will be apparent to persons skilled in the relevant art(s) based on the teachings contained herein. Such alternatives fall within the scope and spirit of the disclosed embodiments.
[053] Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present disclosure. A computer-readable storage medium refers to any type of physical memory on which information or data readable by a processor may be stored. Thus, a computer-readable storage medium may store instructions for execution by one or more processors, including instructions for causing the processor(s) to perform steps or stages consistent with the embodiments described herein. The term “computer-readable medium” should be understood to include tangible items and exclude carrier waves and transient signals, i.e., be non-transitory. Examples include random access memory (RAM), read-only memory (ROM), volatile memory, nonvolatile memory, hard drives, CD ROMs, DVDs, flash drives, disks, and any other known physical storage media.
[054] It is intended that the disclosure and examples be considered as exemplary only, with a true scope and spirit of disclosed embodiments being indicated by the following claims. , Claims:
CLAIMS
WHAT IS CLAIMED IS:

1. A computer-implemented method for dynamic skilling of an Artificial Intelligence (AI) agent (210), the method comprising:
receiving, by the AI agent (210), a first request from a user (212), the first request being indicative of a first technical obstacle requiring diagnostic analysis and troubleshooting;
determining, by the AI agent (210), necessitation of obtaining a first skill;
accessing, by the AI agent (210), one or more data sources (207) to retrieve a primary dataset pertaining to the first skill;
validating the first skill prior to application, validation being performed by the AI agent (210), according to one or more operational parameters, wherein the application is associated with a resolution of the first technical obstacle;
receiving, by the AI agent (210), a second request from the user (212), the second request being indicative of a second technical obstacle requiring diagnostic analysis and troubleshooting;
determining, by the AI agent (210), necessitation of obtaining a second skill;
pruning, by the AI agent (210), the primary dataset prior to accessing the one or more data sources (207) to retrieve secondary dataset pertaining to the second skill;
compiling, via an integration engine (220), pruned primary dataset and the secondary dataset for validation and application of the second skill based on the one or more operational parameters; and
generating a resolution for the second technical obstacle.

2. The computer-implemented method for dynamic skilling of an AI agent (210) of claim 1, wherein the first skill and the second skill correspond to data representation beyond a coverage of a pre-trained data of the AI agent (210).

3. The computer-implemented method for dynamic skilling of an AI agent (210) of claim 1, wherein the one or more data sources (207) are selected from one or more large language models (LLMs).

4. The computer-implemented method for dynamic skilling of an AI agent (210) of claim 3, wherein the accessing of the one or more data sources (207) comprises querying, by the AI agent (210), the one or more LLMs to retrieve the primary dataset and the secondary dataset, and wherein the querying comprises:
generating one or more structured requests based on functional requirements associated with obtaining the first skill and the second skill, and
parsing, by an interpretation module (215), one or more structured requests to extract skill-based attributes for retrieving the primary dataset and the secondary dataset.

5. The computer-implemented method for dynamic skilling of an AI agent (210) of claim 1, wherein the one or more operational parameters include:
performance metrics of the AI agent (210),
response time of the resolution of the AI agent (210),
accuracy of the resolution of the AI agent (210), and
integration efficiency of the integration engine (220).

6. The computer-implemented method for dynamic skilling of an AI agent (210) of claim 1, wherein the validating and the pruning are performed using an iterative moderation technique.

7. The computer-implemented method for dynamic skilling of an AI agent (210) of claim 3, wherein the primary dataset and the secondary dataset includes criteria-based data from the one or more LLMs pertaining to the first skill and the second skill respectively.

8. The computer-implemented method for dynamic skilling of an AI agent (210) of claim 1, the computer-implemented method further comprises transforming the first request and the second request in real-time using one or more data anonymization techniques.

9. The computer-implemented method for dynamic skilling of an AI agent (210) of claim 1, wherein the validation comprises determining performance of the AI agent (210) against pre-defined performance metrics to enable skill certification of the AI agent (210).

10. A system for dynamic skilling of an Artificial Intelligence (AI) agent (210), the system comprising:
a processor (206); and
a memory (204) communicatively coupled to the processor (206), wherein the memory (204) stores processor instructions, which when executed by the processor (206), cause the processor (206) to:
receive, by the AI agent (210), a first request from a user (212), the first request being indicative of a first technical obstacle requiring diagnostic analysis and troubleshooting;
determine, by the AI agent (210), necessitation of obtaining a first skill;
access, by the AI agent (210), one or more data sources (207) to retrieve a primary dataset pertaining to the first skill;
validate the first skill prior to application, validation being performed by the AI agent (210), according to one or more operational parameters, wherein the application is associated with resolution of the first technical obstacle;
receive, by the AI agent (210), a second request from the user (212), the second request being indicative of a second technical obstacle requiring diagnostic analysis and troubleshooting;
determine, by the AI agent (210), necessitation of obtaining a second skill;
prune, by the AI agent (210), the primary dataset prior to accessing the one or more data sources (207) to retrieve secondary dataset pertaining to the second skill;
compile, via an integration engine (220), pruned primary dataset and the secondary dataset for validation and application of the second skill based on the one or more operational parameters; and
generate a resolution for the second technical obstacle.

11. The system for dynamic skilling of an AI agent (210) of claim 10, wherein the processor instructions, on execution, further cause the processor (206) to interface the AI agent (210) with one or more large language models (LLMs) via an application programming interface (API).

12. The system for dynamic skilling of an AI agent (210) of claim 10, wherein the first skill and the second skill correspond to data representation beyond a coverage of a pre-trained data of the AI agent (210).

13. The system for dynamic skilling of an AI agent (210) of claim 10, wherein the one or more data sources (207) are selected from one or more LLMs.

14. The system for dynamic skilling of an AI agent (210) of claim 11, wherein the processor (206) is further configured to access the one or more data sources (207) by:
querying, by the AI agent (210), the one or more LLMs to retrieve the primary dataset and the secondary dataset, and wherein the querying comprises:
generating one or more structured requests based on functional requirements associated with obtaining the first skill and the second skill, and
parsing, by an interpretation module (215), one or more structured requests to extract skill-based attributes for retrieving the primary dataset and the secondary dataset.

15. The system for dynamic skilling of an AI agent (210) of claim 10, wherein the one or more operational parameters include:
performance metrics of the AI agent (210),
response time of the resolution of the AI agent (210),
accuracy of the resolution of the AI agent (210), and
integration efficiency of the integration engine (220).

16. The system for dynamic skilling of an AI agent (210) of claim 10, wherein the processor instructions, on execution, further cause the processor (206) to perform validating and pruning using an iterative moderation technique.

17. The system for dynamic skilling of an AI agent (210) of claim 13, wherein the primary dataset and the secondary dataset includes criteria-based data from the one or more LLMs pertained to the first skill and the second skill respectively.

18. The system for dynamic skilling of an AI agent (210) of claim 10, wherein the processor (206) is further configured to transform the first request and the second request in real-time using one or more data anonymization techniques.

19. The system for dynamic skilling of an AI agent (210) of claim 10, wherein the validation comprises determining performance of the AI agent (210) against pre-defined performance metrics to enable skill certification of the AI agent (210).

20. A non-transitory computer-readable medium storing computer-executable instructions for dynamic skilling of an Artificial Intelligence (AI) agent (210), the computer-executable instructions configured for:
receiving, by the AI agent (210), a first request from a user (212), the first request being indicative of a first technical obstacle requiring diagnostic analysis and troubleshooting;
determining, by the AI agent (210), necessitation of obtaining a first skill;
accessing, by the AI agent (210), one or more data sources (207) to retrieve a primary dataset pertaining to the first skill;
validating the first skill prior to application, validation being performed by the AI agent (210), according to one or more operational parameters, wherein the application is associated with resolution of the first technical obstacle;
receiving, by the AI agent (210), a second request from the user (212), the second request being indicative of a second technical obstacle requiring diagnostic analysis and troubleshooting;
determining, by the AI agent (210), necessitation of obtaining a second skill;
pruning, by the AI agent (210), the primary dataset prior to accessing the one or more data sources (207) to retrieve secondary dataset pertaining to the second skill;
compiling, via an integration engine (220), pruned primary dataset and the secondary dataset for validation and application of the second skill based on the one or more operational parameters; and
generating a resolution for the second technical obstacle.