Real-Time AI Avatar Engineer

LiveX AI, Inc.; Palo Alto (CA); On-site; USD 180,000 - 240,000; Full time

Benefits offered by this job

Competitive salary

Equity

GPU compute access

Job summary

LiveX AI, Inc. is hiring an AI/Machine Learning Engineer to train, fine-tune, and serve generative models end to end. The role focuses on low-latency inference for live streaming avatars and building scalable data, training, evaluation, and deployment pipelines.

You will work with diffusion models, video generation, and neural rendering, collaborating with teams to ship avatar features and iterate on feedback in a fast-paced startup environment.

Qualifications

• 3+ years of hands-on ML engineering or strong research with production-quality code.
• Strong PyTorch and modern DL workflows; ability to reproduce and extend papers.
• Background in diffusion models, video generation, or neural rendering is welcome.

Responsibilities

• Train and fine-tune generative models for talking-head synthesis and avatar video.
• Design data pipelines for large-scale video, audio, and multimodal datasets.
• Optimize models for real-time serving with sub-200ms latency.
• Build and maintain training, evaluation, and deployment infra on GPU clusters using PyTorch, DeepSpeed, Ray, and Kubernetes.
• Define and track quality metrics like lip-sync accuracy and latency.
• Collaborate with backend, product, and research teams to ship avatar features.
• Stay current with generative video and multimodal AI research and apply useful techniques.

Skills

PyTorch expertise

Multi-GPU training

Realtime inference

Video/Avatar ML

Diffusion models

Tools

PyTorch

DeepSpeed

FSDP

Ray

Kubernetes

CUDA

Triton

Triton Inference Server

TensorRT

ONNX

Python

Job description

LiveX AI, Inc. is hiring an AI/Machine Learning Engineer to train, fine-tune, and serve generative models end to end. The role focuses on low-latency inference for live streaming avatars and building scalable data, training, evaluation, and deployment pipelines.

You will work with diffusion models, video generation, and neural rendering, collaborating with teams to ship avatar features and iterate on feedback in a fast-paced startup environment.