Research Scientist

At BRIA, we are at the forefront of Visual Generative AI. Our team of visionaries and innovators

has built a platform that empowers organizations and creative professionals: engineers,

developers, data scientists, and AI experts, to generate and modify visual content safely and at

scale.

Our platform delivers end-to-end visual generation and editing solutions, including foundation

models, APIs, SDKs, and seamless web integration. Built on one of the largest licensed datasets

and developed collaboratively with industry leaders, our technology enables high quality,

copyright protection, and safe content usage.

We’re growing our research group and looking for a Research Scientist who wants to build state-of-the-art models.. If you miss shipping real research papers, models, and open contributions while still working on high-impact product challenges, this is a great time to join. We’re actively releasing new work, and additional models and papers arealready in the oven.

Responsibilities

• Lead research and development of cutting-edge image and video generative AI, including

• text-to-image/video generation, video synthesis, and advanced editing.

• Train and scale large generative models (diffusion / flow-matching and beyond) using

• large datasets and high-throughput training pipelines.

• Own the full research lifecycle: ideation → prototyping → rigorous evaluation →

• deployment-ready solutions.

• Set ambitious long-term research goals while defining and delivering intermediate

• milestones.

• Continuously improve model quality, scalability, and usability through systematic

• evaluation and iteration.

• Stay current with emerging techniques and industry trends to keep BRIA at the SOTA

• frontier.

• Collaborate closely with product managers, engineers, and researchers to define

• requirements and translate research into production.

Requirements

• Degree in Computer Science, Electrical Engineering, or a related field, with first-author

• publications at top-tier venues (e.g., NeurIPS, CVPR, ICML, ICLR, ICCV, ECCV, SIGGRAPH).

• Proven large-scale training experience for image/video generation models (multi-node /

• multi-GPU training, large datasets, strong training/eval discipline).

• 2+ years of hands-on experience training diffusion / flow-matching models.

• 3+ years of experience training, evaluating, and deploying visual GenAI models.

• Strong expertise in image/video representations, processing, synthesis, and evaluation

• methodologies.

Why Join BRIA AI?

If you’re excited by SOTA research and want your work to ship, BRIA offers a rare mix: a strong

research culture with regular paper releases, access to large licensed data, and the ability to turn frontier research into real products. You’ll join a team pushing visual GenAI forward—fast.