Master ML. Build agents. Ship AI. A programme offered by the Continuing Education Programme (CEP), IIT Delhi.
Enquiry about programme
Six months to build real AI systems, from machine learning to autonomous agents - and open doors across AI engineering, research and product roles.
PROGRAMME OVERVIEW
This is not a tools workshop. You will move from the mathematics under modern AI to the architectures defining its frontier: Transformers, fine-tuning, RLHF, RAG, and autonomous AI agents, so you don't just use Generative AI, you can build with it.
Industry vetted curriculum
Six modules, 100+ hours of live sessions plus an optional 2-day campus immersion at IIT Delhi, across four phases. Every module ends in a working artefact: a model or system you can point to, defend, and reuse.
How is this module future-proof
Model architectures turn over every year; linear algebra, probability and the data layer beneath them do not. This is the part of the stack that never needs relearning.
Module 1
Maths foundations: linear algebra & probability · data measures, distributions & estimation.
How this module is helpful
Builds the linear algebra, probability and estimation you need to reason about how models actually learn, rather than treating them as black boxes.
Gets you fluent in the data layer every AI system sits on — SQL, NoSQL and vector databases — before a single model is trained.
Ends in a working data foundation with a business dashboard, so the maths is immediately attached to something you can show.
Want to dive deeper into the details?
Ready to join the Advanced certification in AI: From ML to Agentic AI and take your first step towards success?
PROJECTS YOU WILL SHIP
Six graded builds accumulate into a portfolio, and the programme converges on a live, defended agentic system at Demo Day, IIT Delhi. Capstone topics and tools are indicative. Programme faculty may update them to reflect current industry needs at time of delivery.
A working data layer across SQL, NoSQL and vector stores, with distributions and estimation applied to a real dataset, closing in a storytelling dashboard a business audience can read.
Ships: a queryable data foundation with a business dashboard on top.
Optimisation applied end to end: regression and derivative-based training, then trees, random forests and boosting, plus hierarchical and K-means clustering, each evaluated properly.
Ships: trained regression, ensemble and clustering models with their evaluations.
Deep feedforward networks, CNNs and LSTMs trained on vision and sequence tasks, then a Transformer with attention — with explainable-AI analysis of what the network actually learned.
Ships: trained neural models with an attention and explainability write-up.
A large language model fine-tuned to a task, then grounded in your own corpus through a retrieval-augmented pipeline — embeddings, a vector store and a chunking strategy you can justify.
Ships: a fine-tuned LLM served through a working RAG pipeline.
An autonomous agent built on the perceive-plan-act-learn loop and the ReAct pattern, with function calling and MCP for tool use, short- and long-term memory, and agentic RAG.
Ships: a working agent that uses tools, remembers, and completes a real task.
The reliability layer: a cost-per-task and latency model with caching and routing, guardrails and prompt-injection defence, and trajectory evaluation that scores how the agent reached its answer.
Ships: an evaluation and guardrail harness with a cost and latency model.
WHO SHOULD ENROL
Admission is selective. Candidates are admitted by IIT Delhi Programme Coordinators based on eligibility and a screening review. You will fit if you have a working foundation in engineering, computing, or a technical discipline. No prior AI experience required.
6 Months
DURATION
Online live
FORMAT
100+ hrs
LEARNING HOURS
28th Nov'26
STARTS
01
Core engineering & computing
02
B.Sc / BCA students
03
STEM graduates & professionals
LEARNING OUTCOMES
Every module ends in a working artefact and a capability you can defend. By the end of the programme you can take an AI system from mathematical foundations to a deployed, autonomous agent, and reason about its cost, reliability and safety.
01
Build the data & ML foundation — work maths, SQL/NoSQL and vector data, then train regression, ensemble and clustering models.
02
Train deep neural networks — feedforward nets, CNNs and LSTMs through to Transformers, attention and explainable AI.
03
Build generative models — VAEs, GANs and diffusion, plus LLM architecture, training, fine-tuning and RLHF.
04
Engineer RAG pipelines — prompt engineering and in-context learning, embeddings, vector stores and chunking strategies.
05
Design autonomous agents — perceive-plan-act & ReAct, tool use, MCP, memory and single- / multi-agent orchestration.
06
Ship agents reliably — reason about agent economics, guardrails, prompt-injection defence and trajectory evaluation.
Programme Faculty
Three IIT Delhi faculty lead the programme, working across machine learning, applied AI, and data-driven modelling, from core research to real-world engineering and business systems.
Prof. Manojkumar Ramteke
Research in machine learning, optimization, and evolutionary algorithms, applied to data-driven modelling and AI-based decision systems. Impact · 1,300+ citations · h-index 18
Prof. Hariprasad Kodamana
Ph.D. from IIT Bombay, postdoctoral research at the University of Alberta. Research spans machine learning, deep learning, and data-driven modelling.
Prof. Agam Gupta
Fellow, IIM Calcutta. Works at the interface of technology and business: applied machine learning, business intelligence, and data visualization. Honours · IIT Delhi Teaching Excellence Award
Prof. Manojkumar Ramteke
Research in machine learning, optimization, and evolutionary algorithms, applied to data-driven modelling and AI-based decision systems. Impact · 1,300+ citations · h-index 18
Prof. Hariprasad Kodamana
Ph.D. from IIT Bombay, postdoctoral research at the University of Alberta. Research spans machine learning, deep learning, and data-driven modelling.
Prof. Agam Gupta
Fellow, IIM Calcutta. Works at the interface of technology and business: applied machine learning, business intelligence, and data visualization. Honours · IIT Delhi Teaching Excellence Award
TOOLS AND TECHNOLOGIES
You build in the tools the industry actually uses. Named tools are illustrative of the current industry stack; SQL, NoSQL and vector databases, agent frameworks and evaluation tooling are introduced in the relevant modules and may be substituted at faculty discretion. Tools shown are indicative and may be modified to fit the programme.
WHY NOW
AI has moved from research lab to enterprise core faster than any recent technology cycle, and the frontier has already shifted again, from generative models to autonomous agents. Organisations that can build these systems are pulling away from those that can only buy them, and the engineers who can build them, across ML, GenAI, and agentic AI, are the scarcest resource in the market.
33%
Agentic AI goes mainstream. Share of enterprise software applications set to embed agentic AI by 2028, up from under 1% in 2024.
Gartner
Gartner forecast: 33% of enterprise applications to include agentic AI by 2028 (from <1% in 2024).
3.2:1
The talent runs dry. Global AI demand-to-supply ratio, roughly 1.6M open roles against only ~518K qualified candidates.
Industry estimates
Industry AI-talent demand-to-supply estimates, ~1.6M open roles vs ~518K qualified candidates, 2026.
+280%
Agent builders wanted. Year-on-year growth in agentic-AI job postings in 2026, with average pay near $190K.
Market analyses
Agentic-AI job-posting growth and pay bands, 2026 market analyses.
~40%
A trillion-dollar wave. Compound growth of the generative-AI market, forecast to pass $1.2T by 2034.
Analyst forecasts
Generative-AI market CAGR and 2034 outlook (Fortune Business Insights; Grand View Research). Figures are cited as reported and vary by research firm.
HOW YOU LEARN
A fully live online cadence built around a working week, with every module closing on a working artefact — and an optional campus immersion at IIT Delhi for the capstone and Demo Day.
One connected build from maths foundations and ML through to autonomous agents, not GenAI with agents bolted on at the end.
Every module ends in a working artefact, and the programme converges on a live, defended agentic system at Demo Day.
Taught by named, active IIT Delhi researchers in ML, applied AI and data-driven modelling, not platform-recorded lectures.
A structured career layer — personal branding, business communication, job-search strategy and interview preparation — runs alongside the curriculum, delivered by Varsity by InterviewBit.
An optional two-day campus immersion at IIT Delhi where the capstone agent is built hands-on and presented live at Demo Day.
CAMPUS IMMERSION
This programme is delivered live online, built to fit around a full-time job. An optional two-day campus immersion adds a hands-on capstone build and Demo Day at IIT Delhi, where your agentic AI application is presented live to faculty.
Indian Institute of Technology - Delhi
An optional two-day in-person immersion for the capstone build and Demo Day on the IIT Delhi campus. Note: Travel and accommodation cost will be borne by the learners. IIT Delhi will not be responsible for the same. A hybrid option will be open to candidates who are unable to come on campus.
Build your end-to-end agentic AI application on campus, alongside faculty and your cohort.
Present and defend the working agent live in the room, in front of IIT Delhi faculty.
Build a lasting network with a driven cohort of engineers and working professionals.
Certificate
On completion, participants receive a certificate from the Continuing Education Programme (CEP), IIT Delhi. Two certificate types are issued, based on attendance and assessment performance. A Certificate of Successful Completion is awarded to candidates who score at least 50% marks overall and have a minimum attendance of 50%; a Certificate of Participation to candidates who score less than 50% marks overall and have a minimum attendance of 50%. A verifiable e-certificate only is issued — no printed copy. It reflects the programme title, learning hours, completion status and grade band (where applicable).
Certificate of Successful Completion
Awarded to candidates who score at least 50% marks overall and have a minimum attendance of 50%.
Certificate of Participation
Awarded to candidates who score less than 50% marks overall and have a minimum attendance of 50%.
Admissions
Follow the steps to understand how cohort information works.
Cohort Calendar
Admission Process
Starting at
₹9,897/month
One investment, a career-long return — an IIT Delhi credential, and the applied ML-to-agentic AI skills to put your learning to work from day one.
Payment
Balance + GST as one-shot, or EMI via NBFC partners.
Note on instalments
FAQ
Honest answers to the questions that might be holding you back.
Users write prompts. This programme prepares you to build the systems that answer them.
