Deserialize messy LLM tool arguments into a Rust struct with Laminate
LLM tool-call arguments arrive as messy JSON: numbers as strings, missing fields, invented enum values. Laminate shapes them into a typed Rust struct, and shows every change.
Tagged posts
Laminate: one extract() for every provider's tool calls
OpenAI, Anthropic, and Ollama return tool-call arguments in three shapes. Laminate's provider adapters normalize them in Rust, so one extract() call reads every provider.
Introducing xdg-desktop-portal-generic: The portal backend for the compositors that never had one
On many Wayland compositors, the portal backend that lets apps screen-share, inject input, and read the clipboard is only half there. I built the missing half.
LamBoot: a Linux bootloader that shows its work
LamBoot is a memory-safe UEFI bootloader for Linux that records every trust decision it makes to a log you can read after boot, and recovers from crash loops.
Laminate: the missing layer between messy data and your Rust types
Built on serde, Laminate shapes messy JSON, CSV, and LLM output into typed Rust structs with four coercion levels and a diagnostic for every conversion.
I set out to build an AI document app. I built a PDF parser instead.
I spent months on eighty thousand words of architecture and a 2,700-line Rust PDF parser. The AI document app they were for never opened a window.