Stack

Understand

How the models actually work, from the tokenizer up.

• Attention Mechanisms, from First Principles · 16 lessonsSelf-attention, multi-head, KV cache, MQA, GQA, MLA, FlashAttention — the complete mechanism that makes transformers work.
• The Transformer Architecture · 16 lessonsFrom 'Attention Is All You Need' to GPT-4 and Llama 3 — every layer, every design choice, every scaling decision.
• AI for Everyday Work · 14 lessonsGet real work done with ChatGPT, Claude, and Gemini — writing, meetings, spreadsheets, research — with zero code and a working detector for when AI is wrong.
• How AI Actually Works · 14 lessonsThe plain-English tour of what's inside ChatGPT and Claude — tokens, training, hallucination, RAG, agents — no math, and a map of where to go deeper.
• AI for Marketing & Creators · 14 lessonsContent, ads, and automation for people who ship posts, not code — brand voice, Meta ads, image tools, and connecting AI to the apps you already use.
• AI for Students & Researchers · 14 lessonsStudy so it sticks, find sources that exist, read papers above your level, and know exactly where the line is — plus why fabricated citations happen.
• Tokenization, Inside and Out · 14 lessonsBPE, WordPiece, Unigram, SentencePiece, tiktoken — every algorithm, every tradeoff, every production implication.
• Embeddings, End to End · 23 lessonsOne-hot to Matryoshka, TF-IDF to contrastive fine-tuning, text to audio to graphs — every technique with its formula, every generation with the failure that ended it.

Build

Turning a model into an application that does something useful.

• Building RAG Systems · 23 lessonsGround language models in your own knowledge — retrieval, reranking, evaluation, and production, end to end.
• Claude for Developers · 24 lessonsFrom your first API call to multi-agent systems and the AI-assisted development lifecycle — every Claude surface, tool, and discipline in one course.
• Build AI Tools Without Code · 14 lessonsConfigured assistants, automations that run without you, a chatbot over your own documents, and the guards that make them safe to leave on.
• Your First AI Project · 14 lessonsFrom a blank browser tab to a working tool — your first API call, a loop, structured output, retrieval, and a web page. Minimal code, no setup.
• The Model Context Protocol · 18 lessonsMCP from the wire up on the 2025-11-25 revision — servers, clients, the four primitives, security, sandboxing, remote auth, and a live two-server capstone, with real captured traffic on every page. Its companion, MCP in Practice, builds on the current 2026-07-28 spec.
• MCP in Practice · 36 lessonsBuild MCP servers, clients and apps on the 2026-07-28 protocol and Python SDK v2 — one running system, every snippet executed, and an honest account of what the 2026 spec deleted.
• Prompt Engineering and Context Design · 12 lessonsSystem prompts, structured outputs, chain-of-thought, and the context engineering patterns that make LLMs reliable in production.
• Vector Databases, Inside and Out · 18 lessonsAlgorithms, architecture, and operations — from HNSW internals to production at scale.
• AI Agents in Production · 16 lessonsFrom the agent loop to multi-agent systems — LangGraph, MCP, tool calling, memory, and the guardrails that make agents safe to ship.
• Chunking Strategies for RAG · 18 lessonsFixed-size to late chunking, overlap to context enrichment, evaluation to production lifecycle — every strategy measured, diagrammed, and decided.
• Advanced RAG · 39 lessonsEvery advanced retrieval technique as the answer to a specific failure — query understanding, retrieval loops, graphs, non-prose corpora, and when to stop retrieving altogether.

Ship

Everything between a working prototype and a system you can operate.

• AI for Managers & Team Leads · 14 lessonsWhere AI fits in your operation, how to read a vendor demo, a pilot that can actually fail, and the data policy your team can recite.
• LLM Evaluation and Testing · 27 lessonsllm-evaluation-and-testing
• LLM Fine-Tuning for Production · 14 lessonsLoRA, QLoRA, dataset engineering, and the decision framework for when fine-tuning is the right answer — and when it isn't.
• LLMOps: Serving and Operating LLMs · 17 lessonsModel serving, inference optimization, guardrails, observability, and cost control — the engineering that keeps LLM systems running.

Evolve

The patterns arriving now — agent harnesses, self-improvement, compound systems.

• Loop Engineering and Agent Harnesses · 14 lessonsHarness design, context management, subagent orchestration, evaluation-as-steering, and the Claude Agent SDK — the infrastructure layer that makes agents reliable.
• Autonomous and Headless Agents · 14 lessonsDurable execution with Temporal, event-driven triggers, sandbox isolation, crash recovery, and the governance framework for agents that run without human oversight.
• Self-Improving AI Systems · 13 lessonsAutomated prompt optimization, log-driven harness patching, principle extraction, memory-driven evolution, and the governance framework for systems that modify themselves.
• Compound AI Systems · 13 lessonsMulti-model routing, cascading, the classifier-generator-verifier triad, AI gateways, protocol composition, and the systems engineering that turns models into products.