Tasks:
- * Analyze model outputs and representations
- * Clean video data and design annotation strategies
- * Design experiments and evaluation systems
- * Research implicit 3D and spatiotemporal modeling
- * Train and optimize vision foundation models
Perks/Benefits:
Skills/Tech stack required:
[3D Modeling] [Autoregressive models] [C++11/14] [C++11/14/17] [CI/CD] [CLIP Score] [Consistency evaluation] [CUDA] [DDP] [DeepSpeed] [Diffusion Models] [Distributed Training] [FSDP] [Fvd] [Git] [GPU Acceleration] [Implicit 3D Modeling] [Inception Score] [Latent Dynamics] [Linux] [Megatron] [Python] [PyTorch] [Spatiotemporal modeling] [Temporal consistency] [Temporal consistency evaluation]
Educational requirements:
[Master's Degree]
Role(s):
[Algorithm Engineer] [Computer Vision Engineer] [Engineer] [Learning Engineer] [Machine Learning Engineer] [Vision Engineer] [World Model Algorithm Engineer]