AI Engineer
Embedding VC3 days ago
Palo Alto, CA, USAMid Level
Responsibilities
- Train and fine-tune diffusion, video, multimodal, and other generative models for interactive avatar experiences.
- Build data pipelines for collecting, cleaning, labeling, aligning, and augmenting large-scale video, audio, and multimodal datasets.
- Optimize models and inference pipelines for sub-200ms realtime serving, including distillation, quantization, pruning, caching, streaming inference, and custom CUDA/Triton kernels.
- Build and maintain distributed training, evaluation, and deployment infrastructure on multi-GPU and multi-node clusters.
- Define quality and performance metrics such as lip-sync accuracy, identity preservation, expression naturalness, latency, and throughput.
- Collaborate with backend, product, and research teams to ship avatar features and incorporate user feedback.
- Track research in generative video, multimodal AI, and neural rendering and apply useful techniques to production systems.
Requirements
- At least 3 years of hands-on ML engineering experience, or strong research experience with production-quality code and a record of shipping deep learning models.
- Deep expertise in PyTorch and modern deep-learning workflows, including the ability to reproduce and extend research papers.
- Strong background in diffusion models, video generation, generative AI, computer vision, multimodal audio-vision AI, speech synthesis, or neural rendering.
- Strong Python engineering skills and experience with distributed training across multi-GPU or multi-node systems.
- Proven experience optimizing realtime inference with TensorRT, ONNX, Triton Inference Server, CUDA, mixed precision, or streaming pipelines.
- Preferred experience with talking-head systems, audio-driven animation, lip-sync, digital humans, 3D Gaussian Splatting, NeRF, face reenactment, or avatar synthesis.
- Publications at venues such as NeurIPS, CVPR, ICCV, SIGGRAPH, ICLR, or ICML, or notable open-source contributions, are preferred.
Benefits
- Full-time, on-site work in Palo Alto, California.
- Competitive salary, equity, and benefits; no compensation range is listed.
- Access to substantial GPU compute.
Tech Stack
Categories
About Embedding VC
Embedding VC is an early-stage venture fund that backs generative AI startups, led by current and former builders and operators. The partnership invests capital and hands-on expertise to help AI founders scale, with a focus on progress toward AGI. Founded in 2023 and headquartered in Menlo Park, California, it operates as a private partnership concentrating on AI-native applications and infrastructure.