Embedding VC
Embedding VC
Embedding VC is a Menlo Park early-stage venture capital firm investing at the seed stage and beyond in technology companies built around AI and machine learning. Its portfolio centers on developer and data infrastructure, and the firm backs founders working across engineering, product, and research.

AI Engineer, Generative Video and Avatars

Train and serve generative models for LiveX AI’s interactive digital avatars. Improve multimodal video systems to support real-time streaming with sub-200ms latency.

विवरण

  • Train and fine-tune generative models for talking-head synthesis, audio-driven facial animation, and full-body avatar video
  • Build data pipelines to collect, clean, label, align, and augment large video, audio, and multimodal datasets
  • Optimize models for real-time serving at sub-200ms end-to-end latency using distillation, quantization, pruning, caching, streaming inference, and custom CUDA or Triton kernels as needed
  • Build and maintain training, evaluation, and deployment infrastructure on GPU clusters
  • Set and track quality measures for lip-sync accuracy, identity preservation, expression naturalness, latency, and throughput
  • Work with backend, product, and research teams to deliver avatar features and refine them using user feedback
  • Follow research in generative video, multimodal AI, and neural rendering, and apply relevant techniques to the technology stack

आवश्यकताएँ

  • Bring at least 3 years of hands-on machine learning engineering experience, or strong research experience paired with production-quality code and a record of shipping deep learning models
  • Have deep expertise in PyTorch and current deep learning workflows
  • Be able to read, reproduce, and extend research papers
  • Have a strong background in at least one area: diffusion models, video generation, generative AI, computer vision, multimodal audio and vision AI, speech synthesis, or neural rendering
  • Have solid Python engineering skills
  • Have experience with distributed training across multiple GPUs and nodes
  • Have demonstrated real-time inference optimization experience with TensorRT, ONNX, Triton Inference Server, CUDA, mixed precision, and streaming pipelines
  • Be comfortable working in ambiguous, fast-changing settings and take initiative and ownership
  • Bring a fast-learning startup mindset and eagerness to advance interactive avatar capabilities

लाभ

  • Equity
  • Benefits
  • Access to substantial GPU compute

संबंधित नौकरियाँ

Capital One
Capital One
Capital One
Capital One
Capital One