Tag
The paper introduces LibraryDesignBench, a two-phase benchmark that has one AI agent write software libraries from underspecified requirements, then scores each library by how effectively other agent implementers use it to solve programming problems in minimal code. The evaluation metric combines squared pass rate with solution simplicity relative to a production-library reference.
A blog post exploring a new design for pretty printer implementations in Rust, addressing challenges of adapting functional programming research to a systems language without garbage collection, and comparing existing approaches like the 'pretty' crate and Oppen-style printers.