🛠 Tech Stack

💼 About This Role

You'll design evaluation benchmarks for LTX generative video models, defining what quality means across diverse use cases. You'll be the ultimate authority on model performance, guiding training decisions and ensuring industrial-grade reliability. This role blends research rigor with product intuition to drive the future of video generation.

🎯 What You'll Do

• Evaluate model checkpoints to guide training experiments
• Develop rigorous benchmarks for release candidates
• Build automatic metrics and Reward Models (e.g., for RL)
• Perform root-cause analysis on model failures

📋 Requirements

• Master's or PhD in Computer Vision, ML, or related field
• Deep expertise in evaluation methodology and statistical rigor
• Sharp perceptual intuition for video and audio quality
• Experience with experimental design and significance testing