Module L10
Language models and AI application patterns
Use model APIs with clear output contracts, context management, and measured limitations.
Why this module
Use model APIs with clear output contracts, context management, and measured limitations.
What you produce
A schema-validated model application. Assess the output against the unit outcomes, explanation, safe handling, and independent transfer. Do not require public sharing or use of real sensitive records.
Chapters and units
L10-C01: Models and mechanisms
Explain context and token constraints without claiming comprehension; Choose a model interface for a declared task and budget
- L10-U01 Tokens, embeddings, and transformer intuition Explain context and token constraints without claiming comprehension
- L10-U02 Training, post-training, and model selection Choose a model interface for a declared task and budget
L10-C02: Context and output contracts
Separate trusted instructions from untrusted source content; Reject a schema-valid but unsupported field when evidence is missing
- L10-U03 Prompt and context engineering Separate trusted instructions from untrusted source content
- L10-U04 Structured output and validation Reject a schema-valid but unsupported field when evidence is missing
L10-C03: Interfaces and comparison
Handle errors and record task-level usage; Compare two configurations on fixed examples
- L10-U05 Model interfaces, cost, and streaming Handle errors and record task-level usage
- L10-U06 Comparison and application lab Compare two configurations on fixed examples