Aarik Wells
I spent over a decade turning rigorous structure into something that moves people. Now I do the same thing with code and AI.
I build AI systems that do specific things well. My flagship,MiniBook Suite, is a dual-LLM pipeline I architected end-to-end — it turns a single prompt into a finished, character-consistent illustrated storybook, with a custom-trained Noir style and automated pagination. By day I evaluate AI-generated images and LLM outputs at Outlier.ai — the same precision I trained across 12 years of ballet and a jewellery bench.
CURRENTLY:AI Data Annotator @ Outlier.ai · Google IT Support (in progress) · Toronto / Remote
Systems, pipelines, and utilities built through AI-assisted development.
A dual-LLM pipeline that turns a single idea into a finished, illustrated, character-consistent storybook — automatically.
A custom Model Context Protocol (MCP) bridge connecting Claude AI to Rhinoceros 3D for text-to-parametric design.
A self-contained Windows batch image processing pipeline using a custom AI-directed Rust backend for vector tracing.
A micro-dataset FLUX LoRA trained on photos of a bespoke physical ring to rapidly prototype new jewelry designs.