Agentic AI Is Coming to Enterprise Software. But What Does That Mean for Today's Software Engineers
Ninety-two percent of companies plan to increase their AI investments over the next three years, according to McKinsey & Company. Yet only 1% of leaders describe their organisations as "mature" on the AI deployment spectrum, where AI is fully integrated into workflows and delivers substantial business outcomes. Separately, Gartner forecasts that by 2028, 33% of enterprise software applications will include Agentic AI, up from less than 1% in 2024. These two data points describe the same moment from different perspectives. Organisations are accelerating AI investment, while enterprise software is rapidly moving towards agentic capabilities. At the same time, very few leaders believe their own AI deployments have reached production maturity. The challenge is no longer building another prototype. It is engineering systems that remain reliable, evaluated, secure and observable once they move into production. This is the engineering gap that the iHUB DivyaSampark, IIT Roorkee Post Graduate Certificate in Agentic and RAG Systems for AI Engineering is designed to address. Rather than positioning AI as a collection of tools or experimentation techniques, the six-month live online programme focuses on the engineering discipline required to build, evaluate, secure and operate Retrieval-Augmented Generation (RAG) and Agentic AI systems in production environments. It is intended for technology professionals with intermediate programming familiarity who already build software, manage codebases or lead technology teams, and now need deeper engineering capabilities as enterprises move beyond proof-of-concept AI. What distinguishes the programme is its emphasis on engineering judgement alongside implementation. Across six curriculum pillars and thirteen modules, participants work through the decisions that often determine whether an AI system succeeds in production. The curriculum explores when prompting is sufficient, when RAG engineering is more appropriate, when fine-tuning should or should not be used, and when multi-agent systems introduce unnecessary complexity rather than value. Reliability is embedded throughout the programme, covering evaluation using golden datasets and LLM-as-judge methods, RAG failure analysis, human-in-the-loop workflows with approval gates and audit trails, and prompt-injection defence. A dedicated module also covers MCP, AI security and governance, including retrieval poisoning, tool abuse, guardrails, and GDPR and DPDP compliance. Participants reinforce these concepts through six module-end projects and a capstone that integrates RAG, agent orchestration, model adaptation, deployment and monitoring into one production-oriented system. Approximately 12 hours of live online masterclasses are delivered by IIT faculty, while the programme's weekly live sessions are led by industry experts and programme leaders. The learning opportunity is offered by iHUB DivyaSampark, IIT Roorkee, a Technology Innovation Hub established under the National Mission on Interdisciplinary Cyber-Physical Systems (NM-ICPS), initiated by the Department of Science and Technology, Government of India. Live masterclasses would be conducted by faculty members from IITs and other premier institutions. The programme is intended for software engineers, AI and data professionals, technology architects, product managers and technical leaders who already possess intermediate programming familiarity and want to deepen their ability to evaluate, build and operate production-grade Agentic AI and RAG systems. Participants who successfully complete the programme are also eligible to attend an optional two-day campus immersion at IIT Roorkee's Noida campus. The next cohort begins on 22nd December 2026. The programme runs for six months, requires approximately 8–10 hours per week, and is delivered 100% live online through two live sessions each week, module-end projects and an end-to-end capstone. Upon successfully completing all graded mini projects, achieving a minimum overall score of 50%, scoring at least 50% in the capstone and maintaining at least 50% attendance, participants receive an e-Certificate (Certificate of Completion) issued by iHUB DivyaSampark, IIT Roorkee. For more information on learner benefits, visit ihubiitroorkee.certificates@emeritus.org
Ninety-two percent of companies plan to increase their AI investments over the next three years, according to McKinsey & Company. Yet only 1% of leaders describe their organisations as "mature" on the AI deployment spectrum, where AI is fully integrated into workflows and delivers substantial business outcomes. Separately, Gartner forecasts that by 2028, 33% of enterprise software applications will include Agentic AI, up from less than 1% in 2024.
These two data points describe the same moment from different perspectives. Organisations are accelerating AI investment, while enterprise software is rapidly moving towards agentic capabilities. At the same time, very few leaders believe their own AI deployments have reached production maturity. The challenge is no longer building another prototype. It is engineering systems that remain reliable, evaluated, secure and observable once they move into production.
This is the engineering gap that the iHUB DivyaSampark, IIT Roorkee Post Graduate Certificate in Agentic and RAG Systems for AI Engineering is designed to address. Rather than positioning AI as a collection of tools or experimentation techniques, the six-month live online programme focuses on the engineering discipline required to build, evaluate, secure and operate Retrieval-Augmented Generation (RAG) and Agentic AI systems in production environments. It is intended for technology professionals with intermediate programming familiarity who already build software, manage codebases or lead technology teams, and now need deeper engineering capabilities as enterprises move beyond proof-of-concept AI.
What distinguishes the programme is its emphasis on engineering judgement alongside implementation. Across six curriculum pillars and thirteen modules, participants work through the decisions that often determine whether an AI system succeeds in production. The curriculum explores when prompting is sufficient, when RAG engineering is more appropriate, when fine-tuning should or should not be used, and when multi-agent systems introduce unnecessary complexity rather than value.
Reliability is embedded throughout the programme, covering evaluation using golden datasets and LLM-as-judge methods, RAG failure analysis, human-in-the-loop workflows with approval gates and audit trails, and prompt-injection defence. A dedicated module also covers MCP, AI security and governance, including retrieval poisoning, tool abuse, guardrails, and GDPR and DPDP compliance. Participants reinforce these concepts through six module-end projects and a capstone that integrates RAG, agent orchestration, model adaptation, deployment and monitoring into one production-oriented system. Approximately 12 hours of live online masterclasses are delivered by IIT faculty, while the programme's weekly live sessions are led by industry experts and programme leaders.
The learning opportunity is offered by iHUB DivyaSampark, IIT Roorkee, a Technology Innovation Hub established under the National Mission on Interdisciplinary Cyber-Physical Systems (NM-ICPS), initiated by the Department of Science and Technology, Government of India. Live masterclasses would be conducted by faculty members from IITs and other premier institutions.
The programme is intended for software engineers, AI and data professionals, technology architects, product managers and technical leaders who already possess intermediate programming familiarity and want to deepen their ability to evaluate, build and operate production-grade Agentic AI and RAG systems. Participants who successfully complete the programme are also eligible to attend an optional two-day campus immersion at IIT Roorkee's Noida campus.
The next cohort begins on 22nd December 2026. The programme runs for six months, requires approximately 8–10 hours per week, and is delivered 100% live online through two live sessions each week, module-end projects and an end-to-end capstone. Upon successfully completing all graded mini projects, achieving a minimum overall score of 50%, scoring at least 50% in the capstone and maintaining at least 50% attendance, participants receive an e-Certificate (Certificate of Completion) issued by iHUB DivyaSampark, IIT Roorkee.
For more information on learner benefits, visit ihubiitroorkee.certificates@emeritus.org
