Almanac

Learning path

Newcomer foundations: the tools, the labs, the core ideas

New to AI and wondering where to start? This path introduces the most-used AI assistants, the companies building them, and the key ideas that make them work — in the order that makes each step easier to understand. No background required; just read one after the other.

Beginner7 steps~42 min

7 steps

Begin →

  1. ChatGPT

    Start here: ChatGPT is the tool that brought AI assistants into everyday life, and understanding what it is gives you a concrete anchor for everything that follows.

    Read →BeginnerIn-depth

  2. OpenAI

    Now meet the lab behind ChatGPT — knowing who OpenAI is and what they're trying to do puts the product in context.

    Read →BeginnerIn-depth

  3. Anthropic

    A second major lab with a different philosophy — reading about Anthropic shows that there's more than one approach to building AI.

    Read →BeginnerIn-depth

  4. Claude

    Claude is Anthropic's assistant, so this step connects the lab you just read about to the product it ships.

    Read →BeginnerIn-depth

  5. Reinforcement Learning from Human Feedback

    This is the training technique that turns a raw language model into a helpful, well-behaved assistant — the key step behind both ChatGPT and Claude.

    Read →BeginnerIn-depth

  6. Chain-of-Thought Reasoning

    One of the most important ideas in how modern AI reasons: letting the model think through a problem step by step before answering.

    Read →BeginnerIn-depth

  7. Retrieval-Augmented Generation

    Finish with RAG — the technique that lets AI assistants pull in outside information, bridging the gap between a model's training and the real world.

    Read →BeginnerIn-depth