Senior Ruby on Rails engineer (7+ yrs, FinTech). Building llmtrim - a Rust proxy that cuts LLM bills ~66%. Backend + AI integration.
Native MiniMax-H3 inference acceleration for ComfyUI, integrating Exact Runtime optimizations with Sana Sol-Attn, rectangular Q/KV attention, and composable support for VDN, Spectrum, Untwisting RoPE, Diff-Aid, and Flow mixed-grid workflows.
Local proxy that compresses your LLM API requests so you pay less, with no change to the answers. Trims wasted tokens from prompts, history, tool output, and code before they're sent: -31% input / -74% output, measured live. Any provider, no extra model calls. Also an MCP server and embeddable library (Rust, Python, Ruby, Kotlin, Swift, JS/TS).
💸 Shrink your token bill in herdr: compresses every agent pane's requests (-31% input / -74% output, measured live) and shows the savings on a per-pane badge
Scoop bucket for llmtrim
Swift package for llmtrim: compress LLM API requests to cut tokens, with no extra model calls
Homebrew tap for llmtrim
Production-ready Ruby gem for sending SMS via OVH's http2sms API with Rails integration.
Stop burning tokens. Package manager for AI coding token optimization tools.
A fast, enhanced, and compliant Ruby implementation of the JSON:API specification, evolving from jsonapi-serializer/jsonapi-serializer (originally Netflix/fast_jsonapi) with enhanced features beyond serialization.
Create a JSONAPI Swagger.