Senior Video Foundation Model Researcher

Tencent; Singapore; On-site; SGD 120,000 - 190,000; Full time

Job summary

Tencent Lightspeed Studios is seeking a researcher to drive next-generation video generation foundations. You will design architectures for long video generation, research high-compression tokenizers, and lead large-scale pre-training efforts.

Publication of results and collaboration with the Lightspeed Tech Center will be expected. PhD in AI fields with top-tier publications and strong PyTorch experience are required; experience with distributed training and video modeling is highly preferred.

Qualifications

• PhD in AI-related field with first-author papers at top-tier conferences.
• Experience with diffusion models and autoregressive generation.
• Experience training video/image generation models with PyTorch.
• 3+ years of research experience in video foundation models.

Responsibilities

• Design and iterate on the next-generation video generation foundation architecture.
• Explore core technologies for long video generation, including long-context attention, KV Cache compression, and memory mechanisms.
• Research high-compression-ratio video tokenizers and unified modeling for multi-resolution videos.
• Lead video pre-training at the 10-100 billion token scale, defining data mixture and curriculum learning strategies.
• Publish academic papers and drive technical roadmap decisions.

Skills

Diffusion models

Autoregressive generation

Video generation research

Publications in AI

Leadership

Education

Ph.D. in AI-related fields

Tools

PyTorch

Distributed training

Job description

Tencent Lightspeed Studios is seeking a researcher to drive next-generation video generation foundations. You will design architectures for long video generation, research high-compression tokenizers, and lead large-scale pre-training efforts.

Publication of results and collaboration with the Lightspeed Tech Center will be expected. PhD in AI fields with top-tier publications and strong PyTorch experience are required; experience with distributed training and video modeling is highly preferred.