Mastering Generative AI: Fine-Tuning Transformers

This intermediate-level course addresses the exploding demand for technical generative AI skills, specifically focusing on the highly sought-after ability to fine-tune transformer models. Designed for AI engineers and specialists looking to enhance their career prospects, the course provides comprehensive training in adapting pre-trained language models for specific applications. Students will explore the differences between PyTorch and Hugging Face frameworks while gaining practical experience using pre-trained transformers for various language tasks. The curriculum covers advanced techniques including parameter-efficient fine-tuning (PEFT), low-rank adaptation (LoRA), quantized low-rank adaptation (QLoRA), and model quantization with natural language processing. Through extensive hands-on labs, participants will develop practical skills in loading models, running inference, training with Hugging Face, pre-training large language models, and implementing fine-tuning techniques with PyTorch adapters. The course emphasizes job-ready skills that employers actively seek, enabling graduates to effectively customize transformer models for diverse generative AI applications and significantly enhance their value in the rapidly growing AI job market.