Research Scientist (Video Generation AI)

TL;DR

Research Scientist (Video Generation): Advance multimodal video generation through novel modeling approaches, training objectives, and scaling strategies with an accent on generation quality, controllability, realism in facial expressions, audio synchronization, human motion, and storytelling. Focus on designing compute-efficient low-latency methods, exploring temporal modeling and multimodal alignment, and translating research into product improvements.

Location: In-person required at NYC HQ (Union Square)

Salary: $175K – $275K • Offers Equity

Company

AI-native video platform that intelligently orchestrates production and editing through natural language, leveraging contextual awareness for professional-grade creative decisions.

What you will do

• Develop novel approaches to video and multimodal generative modeling
• Design training objectives, loss functions, and evaluation methods for compute-efficient, low-latency generation
• Explore temporal modeling, controllability, and multimodal alignment
• Conduct empirical studies on scaling behavior and model performance
• Drive rapid experimentation across architectures and training strategies
• Analyze model behavior and translate insights into product improvements

Requirements

• MS/PhD in ML, CS, or related field
• Strong publication record (NeurIPS, ICML, ICLR, etc. or equivalent)
• Deep expertise in generative modeling (diffusion, autoregressive architectures, etc.)
• Deep understanding of transformers and modern multimodal systems
• Experience with large-scale training, empirical research, and real-time inference optimization
• Strong experience with audio representations and audio-visual datasets

Culture & Benefits

• Comprehensive medical, dental, and vision plans
• 401K with employer match
• Commuter benefits, catered lunch multiple days per week, dinner stipend, Grubhub subscription
• Health & wellness perks
• Multiple team offsites per year with monthly team events
• Generous PTO policy