Luma AI is hiring: Technical Artist in New York

Technical Artist This is a rare opportunity to sit at the intersection of frontier research and creative application. As a Technical Artist on the Applied team, you will be the bridge that takes our underlying foundation models and makes them production-ready. You won't just be using tools; you will be defining the workflows, fine-tuning strategies, and aesthetic benchmarks that turn raw model capabilities into magical, controllable experiences for millions of creators. This is a 0-to-1 opportunity to shape the "feel" of Luma's next generation of AI.
Architect and Build Pipelines: Design and implement robust ComfyUI-style pipelines to productionize our models, ensuring they are scalable and efficient.
Own Model Fine-Tuning: Develop and implement LoRA-based fine-tuning strategies to enhance model controllability and steer models toward specific aesthetic directions.
Bridge Research and Design: Collaborate closely with the Research and Design teams to evaluate early model versions, providing the "early eye" for positive or negative trends and translating aesthetic goals into technical requirements.
Drive Model Improvement: Take ownership of identifying model deficiencies and implementing fine-tuning solutions or data enhancements to solve them.
QA for Aesthetics: Work with the Data team to ensure final models meet exceptionally high aesthetic quality standards that match and exceed Luma's product vision.
Prototyping: Rapidly prototype new creative workflows and controllability features to test new research opportunities.
Demonstrable experience with Visual GenAI, either as a creator, fine-tuner, researcher, or engineer.
Strong generalist engineering skills with proficiency in Python. You are comfortable writing scripts and debugging code.
Deep experience with open-source generative AI tools such as ComfyUI, Automatic1111, Kohya, or similar frameworks.
A hobbyist or professional background in art/design leading to a deep understanding of aesthetics and the ability to articulate visual choices using standard artistic language.
Experience with Low-Rank Adaptation (LoRA) techniques and other methods for efficient model fine-tuning.
You are exceptionally motivated and passionate about building amazing models, with a "high agency" drive to solve problems independently.
Active contributions to the open-source Generative AI community (custom nodes, workflows, etc.)
Experience shipping tools that sit between R&D and production.
Background in traditional VFX or Game Development technical art pipelines.
The base pay range for this role is $125,000 – $250,000 per year.

In Summary: As a Technical Artist on the Applied team, you will be the bridge that takes our underlying foundation models and makes them production-ready . You will be defining the workflows, fine-tuning strategies, and aesthetic benchmarks that turn raw model capabilities into magical, controllable experiences . The base pay range for this role is $125,000 – $250,000 per year .

En Español: Artista Técnico Esta es una oportunidad rara de sentarse en la intersección de investigación fronteriza y aplicación creativa. Como artista técnico del equipo aplicado, serás el puente que toma nuestros modelos básicos subyacentes y los hace listos para producción. No solo estarás usando herramientas; definirás los flujos de trabajo, estrategias de ajuste fino y puntos de referencia estéticos que conviertan las capacidades del modelo crudo en experiencias mágicas y controlables para millones de creadores. Esta es una oportunidad de 0 a 1 para dar forma al "sentimiento" de la próxima generación de IA de Luma. Arquitecto y Construir Pipelines: Diseñar e implementar robustas tuberías estilo ComfyUI para producir nuestros modelos, asegurando que sean escalables y eficientes. Own Model Fine-Tuning: Desarrollar e implantar estrategias de ajuste fino basadas en LoRA para mejorar la controlabilidad del modelo y dirigir los modelos hacia direcciones estéticas específicas. Bridge Research and Design: Colaborar estrechamente con equipos de investigación y diseño visual para evaluar las versiones tempranas del modelo, proporcionando el "ojo precoz" para tendencias positivas o negativas y traduciendo objetivos estéticos en requisitos técnicos. Drive Model Improvement: tomar posesión de identificación de estándares de modelo e implementación de soluciones creativas o mejoras a datos genéticos. Usted está cómodo escribiendo scripts y debugging de código. Experiencia profunda con herramientas generativas de IA open-source como ComfyUI, Automatic1111, Kohya o marcos similares. Un entorno aficionado o profesional en arte / diseño que conduce a una comprensión profunda de la estética y la capacidad de articular opciones visuales utilizando un lenguaje artístico estándar.