LLM Academy — Transformers from the inside, training, inference and serving, RAG and fine-tuning, evaluation, security — and what each concept means on AWS, Azure, and Google Cloud.

An interactive, beginner-to-expert curriculum: 70 modules across 10 domains, 10 hands-on tools and simulators, a 202-term glossary.

Domains

Foundations — What a language model is, what next-token prediction really means, and what LLMs can honestly do (6 modules)
Inside the Transformer — Tokens, embeddings, attention, and the block — the machine, opened up (9 modules)
Pre-training — Data pipelines, objectives, scaling laws, and the engineering of a training run (7 modules)
Post-training & Alignment — SFT, RLHF, DPO, and reasoning training — how a base model becomes an assistant (7 modules)
Inference & Serving — Sampling, the KV cache, quantization, batching — where latency and cost are made (8 modules)
Adapting LLMs — Prompting, RAG, fine-tuning, distillation — and how to choose between them (7 modules)
Evaluation — Benchmarks, contamination, judges, and eval harnesses you can defend (6 modules)
Security & Risk — Hallucination, injection, leakage, red teaming — the model-level risk map (7 modules)
LLMs on the Cloud — Bedrock, Azure AI Foundry, Vertex AI, GPUs, cost models, reference architectures (7 modules)
Model Landscape — Model families, open vs closed weights, multimodal and small models, model cards (6 modules)

The learning path — a guided route through every module, beginner to expert.
Tools & simulators — 10 interactive tools.
The Exam Hall — practice drills and mock exams, every answer explained.
Field Notes — practitioner patterns and case studies.
Glossary — 202 terms with primary sources.
Model Release Log — Notable model releases and technique shifts, quarter by quarter — dated and sourced.