Skip to main content
The Human Bit
Ask Bit

Learn

Understand why it worked, one short lesson at a time.

Choose by where you are now, then take one short lesson or bring a real task to Bit.

Never really used it? Try one small thing first (3 minutes).

Bring a real task to Bit →

Choose by state

What best describes you today?

Approximate is fine. Each door leads to one useful next step, not a catalogue.

I am new to AIStart with the plain-English basics, then try one safe move.Start here →I have tried it, but I am unsurePick one stuck work moment and get a safer first move before more lessons.Start here →I want to learn about agentsLearn what an agent can do, what it should not do, and how to check it.Start here →I want to build or use APIsMove into tool calling, context and the first-agent guide without making that the default path.Start here →

Choose your way in.

You have marked 0 of 8 individual pages done outside AI 101. The foundation course has its own lesson progress below.

Build your confidence

Understand what AI is and how to trust it

Tell what these tools really do, pick a first task that is safe to get wrong, and learn to check a result before you rely on it.

AI 101 · 7 lessons · 59 minutes

0 of 7 lessons done

Start lesson 1
See all AI 101 lessons →

Rather than read it all, Bring a first task to Bit →

Use AI for real work

Get more from the tools you already have

Give AI the context it needs, work with its short working memory, and catch a confident answer that is still wrong.

Start here · 0/3 done

How to provide useful context

Learn why useful source material, examples and rules matter as much as the instruction.

See 2 more lessons
  • What a context window really is

    A context window is the model's working memory for one answer, measured in tokens. See why it forgets the top of a long thread, and when to paste, retrieve or summarise.

  • Good-looking is not the same as correct

    A confident answer can still be wrong. An eval is a small, fixed set of pass-or-fail test cases you re-run to catch that. Build the smallest set that would catch your worst failure.

Rather than read it all, Try this on your own work with Bit →

Agents and building

Understand agents, then build one safely

See how an agent takes real actions through tools, where the human checkpoint belongs, and build one small bounded agent before anything real is connected. The concepts you can only see through the API sit here too.

Start here · 0/5 done

Understand agents

See how an agent differs from chat, ordinary software and deterministic automation.

See 4 more lessons
  • Build something controlled

    Create one small Google ADK agent with a bounded tool, clear tests and a human stop point.

  • Why asking the same thing twice can be cheaperFor people who build with the API

    Providers can reuse the stable start of a prompt so repeat requests cost less and answer faster. See what a stable prefix is, what breaks it, and how to order a prompt to keep it.

  • How an AI does things, not just says themFor people who build with the API

    On its own a model only writes text. Tool calling lets it hand your software a precise request to run a real action, then use the result. See what to expose as a tool and what to leave to words.

  • An agent is a model using tools in a loopFor people who build with the API

    An agent is not a smarter chatbot. It plans, calls a tool, looks at the result and decides the next step, until a stop condition. See when a loop earns its keep and where the human checkpoint goes.

Rather than read it all, Talk an agent idea through with Bit →