Prompt Engineering

Why Smarter. Students change a model’s behavior by editing a manifest, with no code, and see the result straight away in the prompt engineering workbench, whose Console shows the configuration, the API calls and every tool call beside the chat. The same prompt can be tried against several providers and models by changing two lines.

Read. Smarter LLM Client, Smarter Plugin, Styling Your LLMClient’s Chat, Smarter API Manifests (SAM).

Exercises.

• Start from the built-in simple LLMClient. Rewrite its defaultSystemRole to give it a persona and a narrow job, then write test prompts that it should answer and prompts that it should decline.

• Run the same ten prompts at defaultTemperature 0 and 1, and on two models. Compare the answers, the tokens used and the cost.

• Give an LLMClient knowledge with a SkillPlugin or a Static Plugin, and the ability to act with a Websearch Plugin or an API Plugin. Discuss when the model decides to use a tool, and how the plugin’s description influences that decision.

• Design an application for a real client (a campus office, a club, a local business), brand its chat with spec.config.styles, and present it.

Course plans. Three complete, ready-to-teach plans, each with a schedule, graded assignments and rubrics, two capstone projects, recommended texts, further study and technical prerequisites. See Course Formats and Grading for what the formats have in common.

• Classroom bootcamp: 8 hours
• Two-week online course: 6 hours
• Four-week online course: 12 hours

See the Smarter Instructor Guide to install Smarter and set up a class.