LLM Fine-Tuning: LoRA, QLoRA & Evaluation
Adapt open-weight language models to your tasks with supervised fine-tuning and parameter-efficient methods, then prove the improvement.
- ✓ Instructor-led live sessions, online or in Bengaluru
- ✓ Hands-on labs in every module
- ✓ Capstone project with trainer review
- ✓ Curriculum customisable for corporate teams
- ✓ Certificate of completion
Next batch: Wed, 25 Nov 2026 · 32 hours · 4 days
Generative AILLM Fine-Tuning: LoRA, QLoRA & Evaluation
Duration: 32 hours · 4 days
Next batch: Wed, 25 Nov 2026
Format: Online live · Bengaluru classroom
Modules: 9 · Topics: 42
Level: Advanced
Fee: ₹14,999
Trusted by leading experts
Upcoming batches
Open batches for individuals; corporate batches on your dates.
| Programme | Starts | Mode | Schedule | Duration | |
|---|---|---|---|---|---|
| LLM Fine-Tuning: LoRA, QLoRA & Evaluation | Wed, 25 Nov 2026 | Online live | Weekday | 4 days | Enquire |
| LLM Fine-Tuning: LoRA, QLoRA & Evaluation | Sat, 09 Jan 2027 | Online live | Weekend (Sat) | 5 Saturdays | Enquire |
| LLM Fine-Tuning: LoRA, QLoRA & Evaluation | Mon, 22 Feb 2027 | Bengaluru classroom | Weekday | 4 days | Enquire |
Soaring demand, and where it takes you
As organisations adopt open-weight models, the ability to adapt them responsibly to domain tasks is a growing specialist skill.
Become an expert
Roles this programme prepares you for:
Become a LLM Fine-Tuning Engineer
Adapts open models to domain tasks and measures the results.
- Decide when fine-tuning beats prompting or RAG
- Build and clean instruction-tuning datasets
- Run supervised fine-tuning with Hugging Face Transformers
About the course
When and how to fine-tune open-weight language models: building instruction datasets, supervised fine-tuning with Hugging Face, parameter-efficient methods like LoRA and QLoRA, and rigorous evaluation to prove the model improved. Includes guidance on when not to fine-tune.
Adapt open-weight language models to your tasks with supervised fine-tuning and parameter-efficient methods, then prove the improvement.
Why should I attend?
ML engineers and data scientists
Applied researchers
MLOps engineers supporting model customisation
Technical leads deciding between RAG and fine-tuning
Prerequisites
Python and PyTorch basics; experience calling LLM APIs; access to a GPU environment (provided in labs)
Level
Advanced
Delivery modes, duration and fee
Online live
Instructor-led virtual classroom; weekday (full-day) or weekend (Saturday) batches
Bengaluru classroom
In-person batches at our Bengaluru training venue
On-site at your office
Private corporate batch anywhere in India, scheduled around your teams
Blended
Live sessions plus guided self-paced labs and mentor check-ins
Duration: 32 hours · 4 days
Course fee (open batch): ₹14,999
Corporate / team batches: per-participant or flat batch pricing on request, based on group size, format and customisation.
What will you learn
- Decide when fine-tuning beats prompting or RAG
- Build and clean instruction-tuning datasets
- Run supervised fine-tuning with Hugging Face Transformers
- Apply LoRA and QLoRA to train efficiently on limited GPUs
- Evaluate task accuracy, regressions and safety
- Package and serve a fine-tuned model
Skills you will gain
Data
Training
Evaluation
Tools and platforms covered
Python
PyTorch
Hugging FaceTransformersPEFT
Weights & Biases
JupyterCourse curriculum
M01 To fine-tune or not
- Prompting vs RAG vs fine-tuning
- Costs and risks
- Choosing a base model
- Open-weight model licensing
- Compute budgeting
🧪 Hands-on: Analyse a use case
M02 Dataset preparation
- Instruction formats
- Cleaning and deduplication
- Synthetic data cautions
- Preference data for alignment
- Train/validation splits
🧪 Hands-on: Build a training set
M03 Supervised fine-tuning
- Training loop basics
- Hyperparameters
- Overfitting signs
- Mixed precision
- Gradient accumulation and checkpointing
🧪 Hands-on: Fine-tune a small model
M04 Parameter-efficient methods
- LoRA
- QLoRA and quantisation
- Merging adapters
- Choosing adapter rank and targets
- Memory estimation
🧪 Hands-on: Train a QLoRA adapter on one GPU
M05 Evaluation
- Task metrics
- Held-out and regression tests
- Safety and bias checks
- Human evaluation protocols
- Comparing against a prompted baseline
🧪 Hands-on: Benchmark base vs fine-tuned
M06 Packaging and serving
- Exporting adapters
- Serving options
- Versioning
- Quantising for deployment
- Model cards and documentation
🧪 Hands-on: Serve the fine-tuned model behind an API
M07 Alignment methods
- Preference optimisation concepts (DPO)
- Reward modelling overview
- When alignment tuning is worth it
- Risks
🧪 Hands-on: Run a small preference-tuning experiment and compare outputs
M08 Distributed and efficient training
- Multi-GPU basics
- DeepSpeed and FSDP concepts
- Experiment tracking
- Cost control
🧪 Hands-on: Track runs with an experiment tracker and compare cost vs quality
M09 Capstone: domain-tuned model
- Dataset build
- Tuning
- Evaluation
- Serving demo
🧪 Hands-on: Fine-tune and serve a domain model with a written evaluation
Certification preparation
This course covers topics that can help participants prepare for the NVIDIA-Certified Professional: Generative AI LLMs (NCP-GENL) exam, which is run by NVIDIA. Check the current exam guide on nvidia.com. Passing is not guaranteed and the exam fee is not included unless stated. [confirm]
Hands-on projects
Domain classifier
Fine-tune a model to classify support tickets, with a before-and-after benchmark.
Style adapter
Train a LoRA adapter for a consistent house writing style.
Why train with us
Built around your stack
Labs and examples adapted to the tools, cloud and data your team actually uses.
Practice-first
Every module ends in a lab, and the course ends with a reviewed capstone.
Practitioner-led
Delivered by Karthick Selvam and vetted specialist trainers from our network.
Visible progress
Attendance, lab completion and assessment results shared with your L&D team.
Certificate of completion
Participants who complete the labs and capstone receive a Learning Dots certificate of completion for LLM Fine-Tuning: LoRA, QLoRA & Evaluation. [confirm certificate wording / sample image]
Learner reviews
“[Verified learner review for this course]”
“[Verified learner review for this course]”
Meet your trainer
Karthick Selvam
Founder · Data Engineering Corporate Trainer, Bengaluru
[Trainer bio, experience and areas of expertise]
Frequently asked questions
Do we need large GPUs?
QLoRA lets many labs run on a single modern GPU; we advise on cloud GPU options.
Can the curriculum be customised for our team?
Yes. Modules, depth, duration and labs can be tailored to your stack and goals.
What delivery formats are available?
Online live instructor-led sessions, Bengaluru classroom, or on-site at your office.
How do we enquire or book?
Use the enquiry form on this page or the Enquire button next to any batch.
What does it cost?
The open-batch fee is ₹14,999 per participant plus GST. Corporate batches are priced on group size, format and customisation.
How long is the course?
32 hours · 4 days. Corporate batches can be split into half-day sessions.
A closer look
About this course
When and how to fine-tune open-weight language models: building instruction datasets, supervised fine-tuning with Hugging Face, parameter-efficient methods like LoRA and QLoRA, and rigorous evaluation to prove the model improved. Includes guidance on when not to fine-tune.
What is LLM Fine-Tuning: LoRA, QLoRA & Evaluation?
Adapt open-weight language models to your tasks with supervised fine-tuning and parameter-efficient methods, then prove the improvement.
Who this course is for
ML engineers and data scientists, Applied researchers, MLOps engineers supporting model customisation, Technical leads deciding between RAG and fine-tuning.
What you will be able to do
Decide when fine-tuning beats prompting or RAG. Build and clean instruction-tuning datasets. Run supervised fine-tuning with Hugging Face Transformers. Apply LoRA and QLoRA to train efficiently on limited GPUs. Evaluate task accuracy, regressions and safety. Package and serve a fine-tuned model.
Career paths
Roles this course supports include LLM Fine-Tuning Engineer, Applied ML Scientist, Model Evaluation Specialist. As organisations adopt open-weight models, the ability to adapt them responsibly to domain tasks is a growing specialist skill.
AI & GenAI guides
Generative AI vs Traditional AI: What Is the Difference?
Traditional AI mostly predicts or classifies. Generative AI produces new content. Both have a place in the enterprise.
What Is Prompt Engineering?
Prompt engineering is the practice of writing instructions that get reliable, useful output from a language model.
What Is ChatGPT and How Does It Relate to Other AI Assistants?
ChatGPT is a conversational AI assistant from OpenAI, built on large language models. It is one of several similar assistants.
How to Compare AI Assistants for Your Team
Several general-purpose AI assistants compete for enterprise use. Rather than chasing rankings, evaluate them against your own needs.