Architecture and Guide to Fine-Tuning and Building Custom LLMs

This meetup session is designed for practitioners planning to build custom models. It assumes familiarity with GenAI and large language models (LLMs), including concepts such as prompt engineering, RAG, agents, fine-tuning, and pretraining. The discussion will cover various AI training stacks, their components, data preparation, experimental tracking, distributed training frameworks, deployment and scaling orchestration, as well as the advantages and disadvantages of different approaches.