The AI Starting Point

Courses / Understanding Language Models and AI Agents

Understanding Language Models and AI Agents

See how an assistant produces its answers, why it makes mistakes, and how agents use tools safely.

6 lessonsAbout 4 hoursBeginner, no coding needed

Progress

Loading…

Not sure yet? Get the free starter lesson first.

Enter your email and we will send it to you.

$40 $50
R680 (about $40)

What you will be able to do

  • Explain tokens, context and prediction in plain words
  • Spot why an assistant gets things wrong
  • Explain retrieval and grounding
  • Describe an agent's think, act, observe cycle
  • Write a small test set for an assistant

What you need: No account is needed. Use the fictional examples; never put secrets or private data into an exercise.

Lessons

  1. 1What is a language model?25 minutes · interactive diagram · Lesson quiz
  2. 2Tokens and context25 minutes · Lesson quiz
  3. 3Prompts, boundaries and untrusted text30 minutes · Lesson quiz
  4. 4Why mistakes happen, and grounding35 minutes · Lesson quiz
  5. 5Models, tools and agents40 minutes · interactive diagram · Lesson quiz
  6. 6Test a small system35 minutes · Lesson quiz

Final project

Choose a fictional FAQ or policy, write an answer-only-from-it prompt, create six tests and record expected against actual results. Explain what the tests prove and what they do not.

Kit included: LLMs and Agents Kit

Includes four hands-on experiments, prompt patterns, a hallucination checklist, an agent design canvas with a worked example, and a safety and test plan. (at no extra price)

Your kit unlocks with this course.

Next steps

  • Take the Codex or Claude Code course to apply the plan, approve, build, test and review habit to a real project.
  • Review AI Foundations if checking and privacy habits need practice.
Start lesson 1