🛠 Tech Stack

💼 About This Role

You'll lead research on generative video and audio models for real-time human simulation, working at a Series A startup backed by Sequoia and YC. Your core impact will be advancing text-to-speech and speech-to-speech models that power expressive AI avatars. This role offers the chance to publish at top venues and see your research productionized.

🎯 What You'll Do

• Lead research on generative audio and video models
• Collaborate with Applied ML team on productionization
• Stay current with latest advancements and drive innovation

📋 Requirements

• Proven experience with flow matching, diffusion models, or auto regressive networks in audio domain
• Experience training deep learning models from medium to large scale
• Experience building streaming text-to-speech or speech-to-speech models
• Publications in top-tier venues like CVPR, NeurIPS, or equivalent

✨ Nice to Have

• PhD or equivalent experience preferred
• Skills in 3D graphics or Gaussian splatting
• Experience leading research teams

🎁 Benefits & Perks

• 🏖️ Unlimited PTO
• 🏥 Competitive healthcare
• 💻 Gear stipends
• 📈 Equity
• 🔄 Flexible work schedule