Get Started
Hunting for you, 24x7Get the iPhone app

Your hunt runs in your pocket, and tells you the moment a strong match opens.

Download on theApp Store
Cantina LabsPosted in November 2026

Research Scientist, Real-Time Video Generation

From EuropeRemoteFull-time
Who we'd reach at Cantina LabsFound and reached by Power Apply
Best leadHiring manager

Leads the team this role joins

Recruiter

In-house, owns the opening

Posted this job

Put the opening up

Power Apply

From your own Gmail and LinkedIn.

About the role

This is a fully remote research scientist role focused on real-time video generation for interactive characters, open to candidates based in Europe. It suits researchers who have trained or fine-tuned video or related generative models and can turn open questions into quick, useful experiments. The employer is a social AI company building real-time models for expressive, lifelike characters. The work spans streaming generation, character motion and speech, fine-tuning, distillation, inference speed, data curation and evaluation.

What you’ll do

  • Explore streaming generation that stays responsive and visually consistent over time.
  • Study how facial expression, gesture and body movement combine into one coherent character.
  • Improve how speech and motion align across different identities and visual styles.
  • Adapt video models through fine-tuning and post-training on new data and behaviors.
  • Make generation faster and more efficient through distillation and runtime improvements.
  • Work through training examples, find coverage gaps and recurring failure modes, and improve curation.
  • Build evaluations that show progress, combining metrics with close inspection of generated video.
  • Read and reproduce relevant research, then share early experiments and demos with the team.
  • Help the team decide which directions deserve more time and compute.

What they’re looking for

  • Experience training or fine-tuning video or related generative models.
  • Ability to learn unfamiliar methods quickly and turn them into useful experiments.
  • Good experimental judgment, ownership of work and clear communication.

What’s on offer

  • Pay depends on skills, experience, job scope, location and market data. No figure is published.
  • Benefits listed for U.S.-based roles: competitive salary and equity, medical, dental and vision cover, 42 days of paid time off, parental leave, fertility support, a 401(k), a lifestyle spending account and meals for in-office staff.
  • Substantial freedom to propose directions and fast iteration with early sharing of results.

Skills

video generationgenerative modelsfine-tuningpost-trainingmodel distillationinference optimizationstreaming generationmodel evaluationdata curationresearch reproduction

Questions about this role

Is the role remote, and can I apply from Europe?

Yes. The role is fully remote and the employer accepts candidates based in Europe.

How much does it pay?

No figure is published. Pay depends on skills, experience, job scope, location and market data.

What experience is needed?

Experience training or fine-tuning video or related generative models, plus the ability to learn new methods quickly.