Research Scientist, Multimodal Video AI

Cantina, Inc.; California; Sur place; USD 200 000 - 320 000; Plein temps

Avantages offerts par ce poste

Equity

Medical, dental, and vision insurance

PTO & holidays (42 days)

Parental leave

401(k)

Lifestyle spending account

Free lunch & snacks

One Medical membership

Résumé du poste

Cantina Labs seeks a research-focused engineer to help build next-generation native video and multimodal foundation models. You will contribute across architecture, data, training, evaluation, and deployment, shaping technical direction from an early stage.

Ideal candidates have strong research experience in generative modeling, hands-on work with large-scale video or multimodal models, and a track record of impactful work through publications or open-source contributions.

Qualifications

• Strong research and engineering experience in generative modeling, including diffusion models, flow matching, DiTs, video generation, multimodal models, or related fields.
• Hands-on experience training and evaluating large-scale image, video, or unified multimodal models using modern deep learning frameworks and distributed training systems.
• Proven track record of developing impactful models or systems demonstrated through research publications, open-source contributions, production impact, or other significant work.
• Ability to independently own ambiguous research problems and move from ideas to experiments in a collaborative environment.
• Specialized depth in one or more areas across the foundation model lifecycle (architecture, data curation, post-training, deployment, etc.).

Responsabilités

• Research, develop, and scale native video and multimodal foundation models from prototypes to large-scale pre-training.
• Explore new architectures, objectives, and conditioning for video generation and multimodal interaction.
• Build data curation, distributed training, evaluation, and post-training pipelines for high-quality generation.
• Design experiments to understand scaling, generation quality, controllability, and inference efficiency.
• Collaborate with researchers, engineers, and product teams to shape roadmap and translate advances into capabilities.
• Contribute to research publications and open-source releases when appropriate.

Connaissances

Generative modeling

Video generation

Multimodal models

Distributed training

Research publications

Collaborative work

Model architecture

Description du poste

Cantina Labs seeks a research-focused engineer to help build next-generation native video and multimodal foundation models. You will contribute across architecture, data, training, evaluation, and deployment, shaping technical direction from an early stage.

Ideal candidates have strong research experience in generative modeling, hands-on work with large-scale video or multimodal models, and a track record of impactful work through publications or open-source contributions.