Program Schedule & Modules
A comprehensive six-day curriculum designed to take you from LLM fundamentals to production-grade deployment
Program Highlights
of intensive instruction combining theory and GPU practice
presenting cutting-edge research in LLMs and CI
with PyTorch, HuggingFace, PEFT, TRL, and vLLM
with direct mentorship from industry instructors
with peers, faculty, and industry leaders
upon successful completion
Daily Structure
Complete 6-Day Schedule Breakdown
Detailed session timings, topics, and instructor lineup for each day.
Foundations of LLM Fine-Tuning
Opening Ceremony & Welcome
Meet the instructors and fellow participants. Program overview and logistics.
Speaker / Lead:Prof. Pallavi NikumbhPlenary: Privacy & Security in LLM Systems
Understanding big data analytics and security considerations in modern AI.
Speaker / Lead:Dr. Debmalya BiswasBreak
Lecture: Transformer Architecture Review & Pretraining Objectives
Deep dive into transformer architectures and foundation model training.
Speaker / Lead:Dr. Kishor BurchundiLunch
Hands-on Lab: Setting Up Fine-Tuning Environment
Install libraries (HuggingFace, PEFT, bitsandbytes). Configure GPU workstations.
Speaker / Lead:Lab InstructorsBreak
Discussion & Reflection
Q&A, group discussion, and day wrap-up.
Speaker / Lead:Dr. Anuradha ThakreParameter-Efficient Fine-Tuning Methods
Lecture: LoRA & QLoRA Deep Dive
Theory and implementation of Low-Rank Adaptation and Quantized LoRA.
Speaker / Lead:Prof. Nikhil R. PalBreak
Fine-Tuning Clinic (Small Groups)
Hands-on debugging and implementation of LoRA/QLoRA on open-source LLMs.
Speaker / Lead:Lab InstructorsLunch
Guest Industry Talk: AI in Healthcare
Real-world applications of fine-tuned models in medical AI systems.
Speaker / Lead:Dr. Aniruddha JoshiBreak
Pair Programming: Fine-Tuning on Domain-Specific Data
Teams implement supervised fine-tuning with instructor guidance.
Speaker / Lead:Lab InstructorsAdvanced Fine-Tuning and Alignment
Lecture: Instruction Tuning & Chat Models
Training models on instruction datasets. FLAN and Alpaca approaches.
Speaker / Lead:Dr. Leelavati NarlikarBreak
Lecture: RLHF & Direct Preference Optimization
Alignment techniques: reward modeling, PPO, DPO, and ORPO.
Speaker / Lead:Dr. Leelavati NarlikarLunch
Case Study: Foundation Models & LLM Landscape
Architecture analysis across LLaMA, Mistral, Phi, and other open models.
Speaker / Lead:Dr. Sachin PawarBreak
Poster Session & Peer Review
Groups present initial project work. Peer and mentor feedback.
Speaker / Lead:All ParticipantsInference Optimization and Quantization
Lecture: Quantization Techniques
INT8, INT4, GPTQ, AWQ, and GGUF formats for efficient inference.
Speaker / Lead:Ms. Mita GawareBreak
Fine-Tuning Clinic: Quantization & Pruning
Apply post-training quantization and pruning techniques to fine-tuned models.
Speaker / Lead:Lab InstructorsLunch
Guest Industry Talk: Efficient LLM Inference
Benchmarking and optimization strategies for inference latency and throughput.
Speaker / Lead:Prof. Dr. Surya DurbhaBreak
Lab Continued: Knowledge Distillation & Pruning
Train smaller student models from larger teacher models.
Speaker / Lead:Lab InstructorsServing, Deployment, and Production Systems
Lecture: Production Serving Frameworks
vLLM, TGI, TensorRT-LLM, and continuous batching strategies.
Speaker / Lead:Dr. Pratap SanapBreak
Lecture: RAG Integration & Cloud Deployment
Building retrieval-augmented generation with fine-tuned models on AWS/GCP/Azure.
Speaker / Lead:Rakshit JainLunch
Team Project Work
Focused work time on capstone projects with instructor support.
Speaker / Lead:All InstructorsBreak
Day Reflection & Planning
Prepare for final presentations. Discuss challenges and next steps.
Speaker / Lead:Dr. Anuradha ThakreCapstone Projects and Final Presentations
Team 1-2 Capstone Presentations
20-minute presentations + 10-minute Q&A. Projects evaluated by industry and academic judges.
Speaker / Lead:ParticipantsBreak
Team 3-4 Capstone Presentations
20-minute presentations + 10-minute Q&A. Projects evaluated by industry and academic judges.
Speaker / Lead:ParticipantsLunch
Keynote: Building AI Ecosystems
From research to productization in the AI industry.
Speaker / Lead:Dr. Yogesh KulkarniClosing Ceremony & Awards
Certificate distribution, best project awards, feedback, and networking.
Speaker / Lead:Program OrganizersLearning Outcomes
- ✓Understand transformer architectures, attention mechanisms, and scaling laws
- ✓Master parameter-efficient fine-tuning (LoRA, QLoRA) on custom datasets
- ✓Implement preference alignment with Direct Preference Optimization (DPO)
- ✓Apply post-training quantization (GPTQ, AWQ, GGUF) without quality loss
- ✓Deploy high-throughput inference engines with continuous batching via vLLM
- ✓Design end-to-end production RAG pipelines and evaluate generation quality
What's Included
- ★Complete notebooks, slides, and code repositories
- ★Daily lunch and refreshments
- ★GPU cloud compute access for training and inference
- ★Networking events and banquet dinner
- ★Direct instructor mentorship during labs
- ★Official IEEE CIS Certificate of Completion
Prerequisites
Make sure you meet the baseline requirements before the summer school begins.
Required Knowledge
- •Proficiency in Python programming and NumPy / Pandas
- •Basic understanding of deep learning and PyTorch tensors
- •Familiarity with neural network training (loss, backpropagation)
Technical Requirements
- •Laptop with modern browser and high-speed Wi-Fi capability
- •Google Colab / Kaggle account or modern terminal access
- •All lab code and GPU runtimes will be provided
💡 Refresher notebooks and environment setup guides will be shared with all registered participants prior to the school start date.
Ready to Join the 6-Day Program?
Seats are limited to ensure personalized attention and lab support. Register today to reserve your place.
Register Now