Spellbrush

Lead bringup, administration, and operations for a large-scale anime AI training GPU cluster. Bridge researchers and bare-metal systems using SLURM, filesystems, networking, and Linux sysadmin skills.

Develops 3D first-person anime adventure game in Unity integrating AI companion powered by in-house LLM for dynamic storytelling and puzzles. Requires passion for anime, prior independent 3D Unity game development, and excitement for generative AI.

Conducts advanced LLM research for post-training, roleplay, and real-time integration into an anime-style 3D adventure game AI companion. Requires deep LLM expertise, anime passion, and ability to build scalable infrastructure on a small onsite team.

Researches and develops motion generation models (text-to-motion, audio-to-motion, facial animation) for real-time AI anime companions in narrative-driven 3D games. Requires hands-on AI model training experience and anime passion; on-site in SF or Tokyo.

Supports Spellbrush Japan’s day-to-day back-office operations across administration, accounting, labor support, and coordination with U.S. headquarters. The role suits a bilingual professional comfortable with broad responsibilities in a fast-paced startup environment.

Designs engaging tactics missions using Unity and in-house tools, including enemies, terrain, and objectives. Collaborates with team, playtests content, and contributes to tooling with C# skills. Requires proven game content creation and shipped game experience.

Builds features for an anime mobile tactics game using Unity and C# as a generalist programmer on a small team. Requires 5+ years experience, shipped Unity games, and interest in applying generative AI research.

Builds scalable data pipelines and infrastructure for AI research, processing petabyte-scale anime data across 10k GPUs. Partners with researchers using distributed systems, big data tools, cloud services, requires 3+ years generalist experience.

Conduct AI research on large-scale diffusion models for anime imagery and other modalities, moving experimental models rapidly into production. The role requires research experience, diffusion-model knowledge, and PyTorch proficiency, with on-site work in Tokyo or San Francisco.

Builds and runs end-to-end ML inference infrastructure for generative AI image models across mobile and web platforms. Requires expertise in distributed systems, Kubernetes, Kafka, Redis, GPUs, and multi-cloud environments.

Builds complex, interaction-heavy front-end applications using React and TypeScript for anime/video game tools, focusing on responsive UIs with Canvas/WebGL. Requires anime passion, design sense, and on-site collaboration in small teams.