A project-based LLM engineer course covering RAG, agents, reasoning models, and image generation, taught live by a best-selling AI author.
The course runs six weeks in a cohort format with live Saturday sessions and Wednesday office hours. Ali Aminian, a best-selling author who has contributed to AI courses at Stanford, leads every session.
You start with how large language models actually work: data collection, tokenization, transformer architecture, and training techniques like RLHF. From there you move into practical adaptation: fine-tuning with LoRA, prompt engineering, and building retrieval-augmented generation systems.
The course closes with a capstone where you pick your own idea, build it using techniques from the course, and get feedback from the instructor directly.
Sessions are recorded. Time commitment is four to six hours per week. Basic Python is needed for the projects but not required to follow the lectures.
What you will learn:
How LLMs are pre-trained and post-trained, including RLHF and reinforcement learning techniques.
How to build RAG systems with vector indexing, chunking strategies, and prompt engineering.
How to build AI agents using tool calling, ReACT, and multi-agent coordination.
How reasoning models scale at inference time using techniques like Tree of Thoughts and self-consistency.
How diffusion models work for text-to-image and text-to-video generation.
Ideal for: Developers and career changers who want structured, project-based AI engineering skills.