
The real shift happening across organisations is from AI assistance to autonomous systems that reason, plan and act independently. Learning Agentic AI is no longer optional for professionals who want to stay relevant, it is the most strategically valued capability in the market today.
The demand for professionals who can build and deploy AI agent systems is growing faster than supply at every level.
The Advanced Certificate in Agentic AI is a 6-month programme priced at INR 1,45,000 plus GST from CEP, IIT Delhi designed to build real-world AI capability for working professionals. It enables you to move from understanding AI to designing and deploying production-grade agentic systems. Through a structured, application-first curriculum, you will work on hands-on labs, mini-projects, and an enterprise capstone, gaining the ability to build autonomous AI workflows using modern tools and frameworks.
As organisations accelerate AI adoption, the gap between experimentation and execution continues to widen. The Advanced Certificate in Agentic AI programme addresses that gap by equipping you to design, deploy, and scale agentic AI systems that drive measurable business outcomes. By the end of the programme, you will be able to move beyond theoretical concepts to building intelligent systems that automate decisions, optimise processes, and create real enterprise impact.

Built for Agentic AI - Not Just AI Usage
Purpose-built for the next wave of AI not a GenAI course with agent modules added on.

100% Live-online Faculty-Led Learning
Every session delivered live-online by IIT Delhi faculty no recorded content, no outdated material.

Apply What You Learn
Every module ends with a hands-on project building a portfolio of real, deployable work as you go.

10+ Industry-Grade Tools
Work with 10+ industry-grade AI tools including LangChain, AutoGen and CrewAI and more

Interactive ML learning with VisualML Lab Pro
An interactive ML training environment for hands-on experimentation beyond traditional classroom settings

IIT Delhi Credential
Earn an e-Certificate from IIT Delhi (CEP) ranked #1 in South Asia in India.
Note:
The primary mode of learning for this programme is via live online sessions with faculty members. Post session video recordings will be available throughout the programme duration. Emeritus or the institute does not guarantee availability of any session recordings.
The programme highlights presented above are approximate, and subject to change depending on the availability and expertise of the teaching faculty, as well as the programme's desired outcomes.
The Advanced Certificate in Agentic AI programme is designed for working professionals who want to move from AI familiarity to building and deploying autonomous AI systems. Learners should have basic Python scripting ability and a conceptual understanding of AI/ML. Prior experience building ML models is not required.
Engineering and Technology Professionals You build with AI today. This programme takes you from implementation to architecture designing and deploying autonomous systems that reason, act, and deliver at scale.
Product, Innovation and Transformation Leaders You drive change. This programme gives you the technical depth to lead Agentic AI initiatives, make informed product decisions and deliver transformation that goes beyond surface-level adoption.
Business & Functional Leaders You set the direction. This programme equips you to understand, evaluate and champion Agentic AI across your organisation turning strategic intent into measurable business outcomes.
The Advanced Certificate in Agentic AI programme curriculum is anchored in the academic rigour and credibility of IIT Delhi, following a structured, end-to-end learning journey, from foundational concepts in agentic AI to the design, orchestration, and deployment of enterprise-grade autonomous systems, enabling real-world application and impact.
What is Agentic AI: concepts, scope, and positioning in the AI landscape
Autonomy, goals, actions, and feedback loops in agent systems
High-level agent architecture: observe → reason → act → learn cycle
Agentic AI vs GenAI, automation, and traditional AI systems
Model Context Protocol (MCP): context management, state handling, and tool contracts
Real-world use cases and industry applicability
Module-end Project 1: Agentic System Design Blueprint
Role of LLMs in agentic decision-making and autonomous reasoning
Reasoning workflows and structured prompting techniques
Advanced prompt engineering and prompt patterns (ReAct, reflection, constraints)
Planning, task decomposition, and self-reflection mechanisms
Reliability, constraints, and failure modes of LLM-based agents
Designing dependable reasoning pipelines for production systems
Cyber-physical systems integration: real-world case studies
Module-end Project 2: Autonomous Prompt Generation & Refinement System
Single-agent vs multi-agent systems: architectural differences
Role-based agents and dynamic task distribution
Communication protocols and coordination mechanisms between agents
System design principles, scalability considerations, and complexity trade-offs
Real-world multi-agent use cases: enterprise workflows, collaborative problem-solving
Module-end Project 3: Multi-Agent Collaboration System
End-to-end agentic workflows: from goal specification to execution
MCP-based orchestration of agents, tools, and external APIs
Enterprise automation and integration with legacy systems
Monitoring, control mechanisms, and human-in-the-loop oversight
Event-driven architectures and real-time agent responsiveness
Module-end Project 4: Agent-Orchestrated Automation Pipeline
Retrieval-Augmented Generation (RAG) for agentic systems
Long-term agent memory architectures and state management
Deployment considerations: cloud, edge, and hybrid architectures
Governance frameworks, risk management, and responsible autonomy
Evaluation metrics, continuous monitoring, and compliance requirements
Enterprise and industry perspectives on agentic AI adoption
Feeds directly into the Capstone Project
Note:
Modules/ topics are indicative only, and the suggested time and sequence may be dropped/ modified/ adapted to fit the total programme hours.
The primary mode of learning for this programme is via live online sessions with faculty members.
Post session video recordings will be made available until the programme duration.
Emeritus or the institute does not guarantee availability of any session recordings.
Design an Agentic System from the Ground Up Create a complete blueprint for a real-world use case, mapping agent behaviour, tools, and system architecture.
Build a Self-Improving AI System Develop a system that generates, tests, and refines its own prompts using advanced reasoning techniques.
Create a Multi-Agent Collaboration System Design agents that work together to solve complex tasks through coordination and communication.
Deploy an End-to-End Automation Pipeline Orchestrate agents, APIs, and workflows into a live automation system with real-time triggers and controls.
Deliver a Full-Scale Enterprise AI System Build and evaluate an end-to-end agentic AI solution solving a real business problem across domains.
The capstone is the culmination of the Advanced Certificate in Agentic AI programme. Participants design, build and evaluate an end-to-end AI agent system that solves a real-world enterprise problem demonstrating autonomy, reasoning, orchestration and responsible deployment.
Project Scope Options:
Customer Support Automation: Intelligent agent handling complex multi-turn conversations with knowledge retrieval
Enterprise Workflow Automation: Multi-agent system coordinating across departments including procurement, approvals and notifications
Research and Analysis Agent: Autonomous system gathering, synthesising and reporting insights across specific domains
Cyber-Physical System Control: Agent-based monitoring and decision-making for IoT and sensor networks
Custom Domain: Participant-proposed application with instructor approval
Capstone Deliverables: Participants conclude the programme by delivering a production-ready agentic AI system that addresses a real-world enterprise problem. The capstone is evaluated on the following deliverables.
GitHub Repository with complete source code and documentation
System Architecture Diagram showing agent design and interactions
Video Demonstration (5-7 minutes) live walkthrough showing the AI agent in action
Final Presentation (15-20 slides) to cohort and IIT Delhi faculty
Note: Placement/Career Support will not be offered for this programme and IIT Delhi will not be responsible for any placement-related assistance.
Dimension | IIT Delhi Agentic AI Programme | Typical Agentic AI Programmes |
Curriculum Depth & Scope | End‑to‑end coverage from agent foundations to enterprise deployment, including multi-agent systems, orchestration, governance, and automation | Focused primarily on prompt engineering, LLM usage, or isolated agent demos |
Focus Beyond GenAI | Strong emphasis on true agentic behaviour: autonomy, goal-driven reasoning, feedback loops, and learning | GenAI‑centric; often treats agents as prompt wrappers or chatbots |
Production Readiness | Designed for real-world, production-grade systems with monitoring, memory, RAG, failure handling, and deployment strategies | Mostly prototype-level implementations with limited production considerations |
Framework & Tool Exposure | Hands-on with 10+ industry-grade tools; LangChain, AutoGen, Docker and more, including VisualML Lab Pro. | Limited tool exposure, often restricted to one framework or API |
Agent Orchestration | Deep focus on Model Context Protocols (MCP), tool contracts, state management, and agent orchestration | Minimal coverage of orchestration or system-level context management |
Multi-Agent Systems | Dedicated module on multi-agent architectures, coordination, communication, and scalability | Multi-agent concepts covered superficially or not at all |

Programme Coordinator and Faculty, IIT Delhi Centre for Sensors, Instrumentation and Cyber-Physical System Engineering (SeNSE)
Prof. Santhosh Sivasubramani works at the intersection of autonomous decision systems, agentic AI, and intelligent computing architectures. These form the foundation of this p...

Understand how agentic AI systems think, act, and learn

Turn LLMs into reliable reasoning engines

Design and deploy single and multi-agent systems

Build end-to-end agentic AI workflows at scale

Develop production-ready agentic AI solutions

Ensure reliable, governed, and responsible AI systems

Build a real-world, enterprise-grade AI system
The IIT Delhi Advanced Certificate in Agentic AI is a 6-month live-online Agentic AI certification programme offered by CEP, IIT Delhi. It is designed for working professionals who want to learn Agentic AI and move from AI familiarity to building and deploying production-grade autonomous AI agent systems in enterprise environments.
IIT Delhi Advanced Certificate in Agentic AI is a 100% live online Agentic AI course. All sessions are delivered live by IIT Delhi faculty every Saturday from 3:30 PM to 6:30 PM with no physical attendance required. Recordings of the live sessions will be made available to enrolled participants, ensuring you never miss a concept even if you are unable to attend a session making it one of the most flexible ways to learn Agentic AI online while working full time.
Unlike a Generative AI course or certification that focuses primarily on prompt engineering and LLM usage, the IIT Delhi Advanced Certificate in Agentic AI programme is built specifically for Agentic AI. It covers autonomous agent systems, multi-agent AI design, workflow orchestration, RAG pipelines and enterprise deployment going well beyond what any Gen AI course covers.
The IIT Delhi Advanced Certificate in Agentic AI programme is open to graduates and diploma holders with a minimum of one year of work experience. Learners should have basic Python scripting ability and a conceptual understanding of AI and ML.
The IIT Delhi Advanced Certificate in Agentic AI programme gives you hands-on experience with 10+ industry-grade AI agent frameworks and tools including LangChain, AutoGen, CrewAI, n8n, Flowise, Hugging Face, Docker and more, the same tools top AI teams use to build and deploy AI agent systems in production. The programme also includes VisualML Lab Pro for interactive ML experimentation.
Yes. The IIT Delhi Advanced Certificate in Agentic AI has Module 3 which is dedicated entirely to designing autonomous multi-agent systems covering agent coordination, communication protocols, role-based design and scalability. Module 4 covers Agentic AI workflow orchestration including MCP-based orchestration of agents, tools and external APIs.
Participants who achieve 60% or more overall with at least 50% attendance receive a Certificate of Successful Completion. Candidates who score less than 60% marks overall or have a minimum attendance of 50%, will receive a Certificate of Participation issued by the CEP, IIT Delhi organised by the Centre for Sensors, Instrumentation and Cyber-Physical System Engineering (SeNSE), IIT Delhi.
All assignments, module-end projects, weekly quizzes and the final enterprise capstone are graded by IIT Delhi faculty ensuring academic rigour and institution-level evaluation standards throughout the IIT Delhi Advanced Certificate in Agentic AI programme.
The capstone is the final milestone of the IIT Delhi Advanced Certificate in Agentic AI programme where participants design, build and deploy an end-to-end AI agent system that solves a real-world enterprise problem. Deliverables include a GitHub repository, system architecture diagram, video demonstration and a final presentation to cohort and IIT Delhi faculty making this a true build AI agents’ course with production-level outcomes.
Participants who successfully complete the programme receive an IIT Delhi e-Certificate issued through the CEP, IIT Delhi is ranked #1 in South Asia by QS World University Rankings 2026 and #2 in Engineering category in India by NIRF 2025 making this one of the most credible AI agent certifications and AI certificate courses in India available today.
The IIT Delhi Advanced Certificate in Agentic AI programme requires approximately 4 to 6 hours per week. This includes the live Saturday session from 3:30 PM to 6:30 PM, short weekly quizzes and time for module-end projects. The programme is designed to be manageable for working professionals without requiring a career break.
Flexible payment options available.
Starts On