Video AI Research Intern: Diffusion & Evaluation

Cantina Labs; Singapore; On-site; SGD 17,000 - 23,000; Part time

Benefits offered by this job

Housing support

Visa support

Stipend

Job summary

Cantina Labs is a social AI company building advanced video generation models and a flagship platform. We are expanding our research lab in Singapore and seek exceptional research interns for a three-month onsite internship in October 2026.

You will work on defined research or engineering problems with a senior mentor from problem formulation through experimentation, evaluation, and potential conference submissions.

Qualifications

• Pursuing a PhD or final-year Master’s in CS/ML.
• Research experience in generative modeling, CV, multimodal learning, or video generation.
• Hands-on with diffusion, flow-based models, distillation, RL, or related post-training techniques.
• Able to formulate hypotheses, design controlled experiments, analyze results, and communicate conclusions.

Responsibilities

• Develop distillation methods for large-scale diffusion/video models.
• Explore inference cost reduction while maintaining quality.
• Develop reward models and preference-based optimization methods.
• Study base-model behavior and inform model development.

Skills

Python

PyTorch

JAX

Research experience

Education

PhD candidate or final-year master's in computer science or ML

Tools

Diffusion models

Model distillation

Reinforcement learning

Job description

Cantina Labs is a social AI company building advanced video generation models and a flagship platform. We are expanding our research lab in Singapore and seek exceptional research interns for a three-month onsite internship in October 2026.

You will work on defined research or engineering problems with a senior mentor from problem formulation through experimentation, evaluation, and potential conference submissions.