LLM Zero to Mastery

LLM Zero to Mastery is my hands-on learning repository for large language models and practical AI engineering.

The repository collects experiments, notes, and implementations around local RAG, transformer fine-tuning, instruction tuning, and model evaluation. I use it as a structured place to turn new concepts into runnable code.

Topics

• Local Retrieval-Augmented Generation over PDF and document collections.
• Vector database creation and document retrieval.
• Fine-tuning GPT-style models with Hugging Face tooling.
• Instruction tuning and instruction-response dataset preparation.
• Practical model testing, evaluation, and reuse.

Technologies

Python, PyTorch, Hugging Face Transformers, vector search, RAG, and Jupyter notebooks.