Programme Parameters
Graduate / Intermediate
3 Days (4.5 Hrs)
Online (e-LMS)
• A basic understanding of the subject area and fundamental programming or scientific concepts.
• A laptop or desktop with a stable internet connection.
• Willingness to complete assignments and the capstone project.
About This Course
Large Language Models (LLMs) and Generative AI takes you from how these systems work to building with them. You learn the transformer architecture and attention that underpin modern LLMs, how models are pretrained and aligned, and why they behave as they do — including their failure modes. The course is hands-on with the practical toolkit: effective prompting, retrieval-augmented generation to ground answers in your data, fine-tuning and parameter-efficient methods, and evaluating generative output. You finish able to design and build a real LLM application with a clear understanding of what is happening under the hood. A verified e-Certificate of competency and e-Marksheet from the Deep Science & Technology Consortium.
🎯 Program Aim
This course covers large language models and generative AI — how transformers and LLMs work, prompting, fine-tuning, retrieval-augmented generation and building real generative applications.
📋 Course Objectives
1. Explain the transformer architecture and attention.
2. Understand pretraining, alignment and LLM behaviour.
3. Apply effective prompting and retrieval-augmented generation.
4. Fine-tune models with parameter-efficient methods.
5. Evaluate and deploy generative-AI applications.
👥 Who Should Enroll?
• Developers and data scientists building with LLMs
• Engineers adopting generative AI
• Technical product teams
• Students of modern AI
🚀 Key Learning Outcomes
• The ability to build a real LLM application.
• A working generative-AI project.
• A grounded understanding of how LLMs work.
• A verified e-Certificate of competency and e-Marksheet from the Deep Science & Technology Consortium.
💎 What You'll Gain
Live & Recorded Sessions
Lifetime access to class recordingse-Certificate on Completion
Cryptographically verified credentialPost-Programme Support
Direct access to mentors & councilHands-On Experience
Notebooks, real-world code & datasetsCurriculum Outline
How Transformers Work
• Tokenisation, embeddings and self-attention explained without the mathematics
• Pretraining, instruction tuning and preference optimisation as three stages
• Context windows, and why a model forgets material inside its own window
Getting Reliable Output
• Instruction structure, few-shot examples and output format constraints
• Chain-of-thought and its uneven benefit across task types
• Systematic evaluation instead of judging a prompt from three examples
Grounding a Model in Your Data
• Chunking, embeddings and vector stores in a RAG pipeline
• Retrieval quality as the usual bottleneck, not the generation step
• Hybrid search, reranking and citation so an answer can be checked
Fine-Tuning and Its Alternatives
• LoRA and parameter-efficient methods against full fine-tuning
• When fine-tuning helps format and tone but cannot add knowledge
• Dataset construction and the cost of getting training examples wrong
Building Something That Holds Up
• Evaluation sets, regression testing and LLM-as-judge with its biases
• Latency, cost per request and caching as design constraints
• Hallucination, prompt injection and guardrails at the application boundary
Technical Specifications
| Parameter | Requirement |
|---|---|
| Covered Tool / Platform | Python |
| Covered Tool / Platform | NLTK |
| Covered Tool / Platform | spaCy |
| Covered Tool / Platform | Hugging Face Transformers |
| Covered Tool / Platform | Gensim |
| Covered Tool / Platform | BERT |
| Covered Tool / Platform | GPT |
Frequently Asked Questions
Scholar Feedback & Reviews
Based on 0 scholar submissions
Rating Breakdown
No verified reviews published yet. Be the first to share your academic experience.
Leave Scholar Feedback
Your rating will help prospective scholars. Ratings below 3 stars are routed privately to the faculty mentor for immediate response.
