Research Scientist

Job Overview

As a Research Scientist, you will investigate technically ambitious problems at the intersection of generative modeling, computer vision, computer graphics, multimodal systems, and professional design workflows.

The position includes room for academic publication and longer-horizon research while remaining connected to projects with meaningful product impact. Research may span areas such as diffusion models, transformers, video generation, image editing, 3D generation, neural rendering, controllable generation, vector graphics, design automation, and multimodal representation learning.

What You’ll Do

• Conduct original generative AI research for professional creative workflows spanning image, video, 3D, vector graphics, and design.
• Develop new methods that advance the state of the art while remaining useful for real creative and production workflows.
• Publish research at leading academic venues where appropriate.
• Contribute research expertise to high-impact internal projects.
• Explore multimodal generative systems that connect image, video, 3D, vector graphics, and design.
• Investigate approaches that improve controllability, editability, and reusable asset creation.
• Translate research findings into prototypes, experiments, and model capabilities that can help inform product direction.
• Collaborate with researchers, engineers, designers, and product teams to identify research problems with meaningful downstream impact.
• Design datasets, benchmarks, and evaluation protocols for professional creative workflows.
• Evaluate areas including output quality, editability, consistency, brand adherence, and production usefulness.
• Stay current with frontier research in generative media and identify opportunities to combine ideas from computer graphics, vision, language, video, and creative tooling.
• Contribute to a research environment focused on technical depth, intellectual breadth, publication quality, and practical relevance.

Research Areas

• Generative image models.
• Video generation and video editing.
• 3D asset generation and reconstruction.
• Neural rendering and geometry processing.
• Differentiable and inverse rendering.
• Materials, lighting, and relighting.
• Vector graphics and structured visual assets.
• Layout, typography, and design systems.
• Multimodal representation learning.
• Image editing and controllable generation.
• Diffusion models and transformers.
• Professional creative and design automation.

Required Qualifications

• Strong academic research track record in computer vision, computer graphics, machine learning, generative modeling, computational imaging, visual computing, or a related discipline.
• PhD or equivalent research experience in Computer Science, Machine Learning, Computer Graphics, Computer Vision, or a closely related field.
• Demonstrated ability to conduct original research and publish research papers.
• Ability to turn research ideas into working systems or prototypes.
• Experience with generative or multimodal modeling.
• Strong programming skills in Python and PyTorch.
• Ability to implement, adapt, and extend modern machine-learning research code.
• Ability to evaluate both research quality and practical workflow quality.
• Understanding of factors including controllability, iteration speed, editability, consistency, reliability, and integration with existing creative processes.
• Strong communication skills and the ability to explain research ideas to both research-focused and product-focused collaborators.
• Creative technical judgment and an ability to combine ideas from different research areas into new directions.

Generative & Multimodal Experience

Relevant research experience may include one or more areas such as diffusion models, transformers, 3D generation, neural rendering, inverse rendering, video generation, image editing, controllable generation, or design automation.

Stability AI is particularly interested in researchers with broad technical backgrounds who can connect ideas across multiple modalities rather than focusing exclusively on a single generation problem.

Highly Desirable Experience

• Publications at leading research venues such as CVPR, ICCV, ECCV, SIGGRAPH, SIGGRAPH Asia, NeurIPS, ICLR, ICML, or related venues.
• Research experience spanning multiple modalities, particularly image, video, 3D, vector graphics, design, or multimodal generation.
• Experience with professional creative domains such as VFX, games, animation, product visualization, advertising, graphic design, or 3D asset production.
• Experience with controllable generation.
• Experience with layered image editing.
• Experience with material and lighting decomposition.
• Experience developing multi-view consistency techniques.
• Experience with relightable assets.
• Experience with vector-aware generation or structured design outputs.
• Experience creating research prototypes that have influenced product direction or enabled new product capabilities.

Professional Creative Workflows

A central focus of this role is developing AI systems that fit the needs of professional creators and enterprises. The research goes beyond generating isolated images or media outputs and considers how assets can be controlled, edited, refined, structured, reused, and integrated throughout a larger creative process.

This includes thinking about research metrics alongside practical workflow considerations such as consistency, reliability, production usefulness, iteration speed, editability, and adherence to creative or brand requirements.

Research & Product Collaboration

You will work across research, engineering, design, and product functions. Research insights may become experiments and prototypes that help determine future model capabilities and product direction, while practical creative-workflow challenges can inform new research questions.

The position therefore suits researchers who value academic-quality investigation but also want their work to contribute to real-world generative AI systems and professional creative tools.