Subject
2 entries
Structured Outputs
Bookmarks
Experimental Prompting DSL: Origin of Instructor
Jason Liu's experimental prompting DSL for structured LLM outputs via OpenAI function calling — an early version of what became the Instructor library. Shows the origin of the pydantic-based approach to reliable structured extraction from LLMs.
Guardrails: Adding Structure and Validation to LLM Outputs
Guardrails adds structural and semantic validation to LLM outputs — defining schemas for what the model should return and automatically retrying or correcting when outputs don't conform. An early solution to LLM output reliability before JSON mode and structured outputs became standard.