IIT Delhi announces 6-month certificate course in agentic AI systems for graduates, working professionals

The Indian Institute of Technology Delhi (IIT Delhi) offered through its Continuing Education Programme (CEP) has announced a six-month Executive Certificate in Agentic AI Systems. Classes will start on January 9, 2027, and will be held every Saturday from 10 am to 1 pm. The last date to apply is October 31, 2026.

Graduates from science, engineering and commerce disciplines can apply, who have at least one year of professional experience. The IIT Delhi course is designed for AI and data professionals, software developers, technology leaders, IT practitioners and graduates who want specialised training. The fee for this course is Rs 1,45,000 + 18 percent taxes.

The admission process includes application screening and a faculty interview. The learning format will be live faculty-led classes with self-paced resources, with an optional campus immersion. CEP, IIT Delhi, will award e-certificates to participants who have the required attendance and assessment requirements.

Also read At 7 IITs, 1,495 reserved SC, ST, OBC faculty posts vacant; IIT Kharagpur alone has 710, reveals RTI

IIT Delhi Agentic AI course: Curriculum, practical training

The curriculum will have eight modules covering agentic AI fundamentals, prompt engineering, autonomous agent design, enterprise knowledge systems, multi-agent coordination, AI security and governance, deployment and large-scale operations.

Participants will learn about large language models (LLMs), Retrieval-Augmented Generation (RAG) and tools used to develop AI-powered workflows. The course will also introduce frameworks and protocols, including Model Context Protocol (MCP), LangGraph, CrewAI and AutoGen.

A dedicated module on AI Trust, Risk and Security Management (AI TRiSM) will cover model governance, explainability, regulatory compliance and security considerations when deploying autonomous systems.

The programme will include practical assignments, beginning with individual AI assistants and information-retrieval applications before progressing to workflows involving multiple AI agents. Participants will complete a final project to demonstrate their ability to develop and deploy an agentic AI application while addressing security and reliability requirements.