Senior Video Generation & Post-Training Research Scientist

sail technology private limited; Singapore; In loco; SGD 120.000 - 180.000; Tempo pieno

Descrizione del lavoro

Sail Technology Private Limited is seeking a research-focused professional in Singapore to advance video generation, enhancement, and alignment. The role emphasizes post-training methods, diffusion-based models, and collaborations across research and engineering teams.

The ideal candidate holds a Master’s in CS/AI/CV with 2+ years of hands-on experience, including first-author papers, and a strong foundation in diffusion models, RLHF, and evaluation pipelines.

Competenze

• Master’s degree or higher in CS/AI/CV or related field.
• ≥2 years of full-time research in CV and generative AI with video experience.
• First-author papers at CVPR/ICCV/ECCV/ICLR/NeurIPS on video topics.
• Experience with diffusion-based generative models for visual content.
• Familiar with post-training, fine-tuning, and parameter-efficient methods.
• Experience designing video quality evaluation pipelines.
• Strong programming and experimental skills; able to work independently and in teams.

Mansioni

• Design and develop post-training approaches for video generation models.
• Explore video enhancement, super-resolution, restoration, editing.
• Develop datasets for instruction and preference alignment; RLHF, DPO, GRPO, PPO.
• Build evaluation pipelines for fidelity, temporal consistency, and video quality.
• Advance controllable long-video generation and camera motion control.
• Investigate model distillation and acceleration for diffusion-based models.
• Collaborate with researchers, engineers, and product teams.

Conoscenze

Video generation

Generative AI

Computer vision

Diffusion models

Research experimentation

Publication track record

Python programming

Formazione

Master's degree in CS/AI/CV

Strumenti

PyTorch

TensorFlow

RLHF

Model distillation

Experiment tracking

Descrizione del lavoro

Sail Technology Private Limited is seeking a research-focused professional in Singapore to advance video generation, enhancement, and alignment. The role emphasizes post-training methods, diffusion-based models, and collaborations across research and engineering teams.

The ideal candidate holds a Master’s in CS/AI/CV with 2+ years of hands-on experience, including first-author papers, and a strong foundation in diffusion models, RLHF, and evaluation pipelines.