Welcome to my Documentation Hub
This site serves as a central technical repository for my work and research in AI/ML, LLM fine-tuning, and Robotics. Here, you will find detailed implementation guides, technical specifications, and architecture diagrams for various specialized projects.
🚀 Featured Projects
Explore the documentation for my key technical implementations:
Gemma 3-1B Fine-Tuning Project
A complete lifecycle demonstration of fine-tuning the lightweight Gemma 3-1B model for Python code generation using Unsloth and LoRA.
RAGv2 (Retrieval-Augmented Generation)
A high-performance, GPU-accelerated local RAG system designed for private, semantic search over PDF documents using PyTorch and Ollama.
Driver Drowsiness Detection System
A real-time computer vision project using YOLOv11 to detect driver fatigue and prevent accidents through intelligent eye-state classification.
Llama.cpp on Windows Guide
A comprehensive guide for setting up and running Large Language Models locally on Windows environments using various hardware backends.
👤 About the Author
I am Prathmesh Nikam, an AI/ML Engineer passionate about bridging the gap between advanced software intelligence and hardware execution. My work focuses on efficient model training (QLoRA), edge AI optimization, and building robust autonomous systems.