Tool calling and chain-of-thought. Eight questions, 80% to pass. Take your time — these patterns underpin the rest of the track.
| Lesson | Concept |
|---|---|
| 1 | Why tools — LLM that acts vs LLM that just generates text |
| 2 | Tool schema — name + description + JSON Schema parameters |
| 3 | Single tool call — model picks the tool, Python executes, model finishes |
| 4 | Multi-step tool calls — 2 tools in sequence on (3 + 4) * 5 |
| 5 | Chain-of-thought — "let's think step by step" |
| 6 | Few-shot CoT — show 2 examples with reasoning, model copies the format |
Week 2: chained prompts, output validation, self-critique, moderation, eval criteria.
Create a free account to get started. Paid plans unlock all tracks.
Tool calling and chain-of-thought. Eight questions, 80% to pass. Take your time — these patterns underpin the rest of the track.
| Lesson | Concept |
|---|---|
| 1 | Why tools — LLM that acts vs LLM that just generates text |
| 2 | Tool schema — name + description + JSON Schema parameters |
| 3 | Single tool call — model picks the tool, Python executes, model finishes |
| 4 | Multi-step tool calls — 2 tools in sequence on (3 + 4) * 5 |
| 5 | Chain-of-thought — "let's think step by step" |
| 6 | Few-shot CoT — show 2 examples with reasoning, model copies the format |
Week 2: chained prompts, output validation, self-critique, moderation, eval criteria.