Pick a concept and explain it out loud in your own words. An AI tutor listens, tells you what you nailed and what was fuzzy, and asks one follow-up to push you deeper — the Feynman technique, automated.
Phase 1 · Foundations Tokenization Explain how tokenization works and why it matters for LLMs.
Phase 1 · Foundations Transformer Architecture & Attention Explain the transformer architecture, focusing on self-attention.
Phase 1 · Foundations Inference Optimization: KV Cache & Quantization Explain the KV cache and how it speeds up LLM inference.
Phase 2 · Prompting Prompt Engineering Fundamentals Explain the core principles of effective prompt engineering.
Phase 2 · Prompting Structured Output & JSON Mode Explain how to get reliable structured (JSON) output from an LLM.
Phase 2 · Prompting Function Calling & Tool Use Explain how function calling / tool use works with an LLM.
Phase 3 · Embeddings & RAG Embeddings Explain what embeddings are and what makes them useful.
Phase 3 · Embeddings & RAG Retrieval-Augmented Generation (RAG) Explain how Retrieval-Augmented Generation works and why it's used.
Phase 3 · Embeddings & RAG Vector Databases Explain what a vector database does and how it searches.
Phase 3 · Embeddings & RAG Retrieval Strategies: Hybrid Search & Reranking Explain hybrid search and reranking, and why they beat plain vector search.
Phase 3 · Embeddings & RAG Chunking Strategies Explain why and how documents are chunked for RAG.
Phase 4 · Agents Agent Architectures: ReAct Explain the ReAct agent pattern.
Phase 4 · Agents Multi-Agent Systems Explain how and why you'd build a multi-agent system.
Phase 5 · Long-Term Memory Agent Memory Explain the types of memory an agent uses and why.
Phase 5 · Long-Term Memory Context Engineering Explain context engineering and how it differs from prompt engineering.
Phase 6 · LangChain & LangGraph LangGraph Explain what LangGraph is and what problem it solves.
Phase 7 · Evals & Safety LLM-as-Judge Explain the LLM-as-judge evaluation approach and its pitfalls.
Phase 7 · Evals & Safety RAG Evaluation Explain how you evaluate a RAG system.
Phase 7 · Evals & Safety Hallucination Detection & Mitigation Explain why LLMs hallucinate and how to reduce it.
Phase 8 · Ship to Production Cost Optimization Explain how you optimize the cost of an LLM application.
System Design Foundations ACID Properties Explain the ACID properties of database transactions.