Visual Artist & Computer Nerd
I make hybrid workflows that mix hand-painted art with technical engineering. My approach is defined by self-reliance: I build my own workstations, manage my own servers, and create my own animation rigs that solve tricky technical challenges.
I have worked on the Veritasium YouTube Channel for many years, where I have created animations for some of their most complex scientific topics.
I also co-founded an animation studio with my very good friend Alondra Vitae, Mosscrystal Studio. This is where I currently take client work so if you're interested please check us out!
Tools
Skills
More about my work
๐ฅ๏ธ Technical Animation & Rigging ๐ฅ๏ธ
Instead of moving pixels myself, I build systems that move them for me. I specialize in math-based rigging and procedural animation, creating characters and environments that respond dynamically rather than relying solely on keyframes. This includes physics visualization and complex data storytelling.
๐จ My "Craft & Code" Aesthetic ๐จ
My core aesthetic involves integrating analog texture into digital spaces. I engineer workflows that bridge the gap: allowing hand-painted assets (watercolor, gouache) to inhabit 3D environments, or conversely, processing vector data to emulate organic media. From distinct scanning pipelines to procedural stylization, I develop custom software to handle the entire spectrum.
๐ ๏ธ Tool Development ๐ ๏ธ
I treat the studio pipeline as a product itself. I write custom tools (Python, Web Apps) to bridge software incompatibilities, automate repetitive tasks, and optimize asset management. If a tool doesn't exist to solve a specific artistic problem, I build it.
๐ค AI & Machine Learning ๐ค
I maintain a skeptical, utility-focused relationship with AI. I prioritize human-created art and use Machine Learning only as an assistive tool to speed up repetitive tasks.
All models run offline on my own hardware to ensure data privacy and ethical control.
- Computer Vision: Depth estimation and rotoscopy for compositing.
- Audio: Text-to-Speech and Speech-to-Text for transcription and placeholders.
- Code: LLMs for syntax assistance and boilerplate generation.