Research Scientist, Post-Training — Video Generation

The role

You will own RL-based post-training for video diffusion/flow-matching models and develop state-of-the-art reward models. You will also lead post-training evaluation using human and automated metrics.

What you will do

• Run RL post-training for video diffusion/flow-matching models at multi-node scale
• Build video reward models
• Define target evaluation dimensions
• Design and configure preference data collection workflows
• Train and validate learned judges
• Safeguard against reward hacking
• Own post-training evaluation including human preference studies
• Distill RL-tuned models to efficient few-step samplers

Required qualifications

• 2+ years hands-on research experience in post-training or generative modeling
• RL or preference-optimization experience on generative models with evidence of model improvement
• Strong grounding in diffusion or flow-matching models
• PyTorch
• Multi-node distributed training

Nice to have

• Experience developing reward models for visual generation, including VLM-as-judge or large-scale preference data collection
• Distillation expertise (distribution matching, consistency, adversarial approaches), ideally for video models
• Familiarity with video-specific failure modes: temporal drift, motion and physics realism

Benefits

• Competitive salary
• Substantial equity
• Full health benefits
• 401k matching

About the company

Pika empowers creators by building state-of-the-art agentic and multimedia platforms. Their vision is to break down technical barriers to creativity.