- Palo Alto, California, United States
- On-site
- Full-time · Mid
- 3+ years
Help train, fine-tune, and serve the generative models behind realtime, interactive AI avatars that can see, listen, and respond with natural speech and expressive motion.
Compensation
Access to GPU compute for training and experimentation, plus competitive salary, equity, and benefits.
Responsibilities
- Train and fine-tune generative models for talking-head synthesis, audio-driven facial animation, and full-body avatar video
- Build data pipelines for large-scale video, audio, and multimodal datasets
- Optimize models for realtime serving, targeting sub-200ms end-to-end latency, using techniques such as distillation, quantization, and custom kernels
- Build and maintain training, evaluation, and deployment infrastructure on GPU clusters
- Define and improve metrics such as lip-sync accuracy, identity preservation, expression naturalness, latency, and throughput
Requirements
- Nice to have: Talking-head systems
- Nice to have: Lip-sync systems
- Nice to have: 3D Gaussian Splatting
- Nice to have: NeRF
- Nice to have: Face reenactment
