Machine Learning Engineer, Video

Cantina

full-time

Posted on: 9/4/2025

Location: California • 🇺🇸 United States

✨ AI Apply

💰 $175,000 - $225,000 per year

JuniorMid-Level

AWSCloudDockerDynamoDBKubernetesPythonPyTorchSQLTensorflow

About the role

Own model deployment end-to-end – take video AI models from research to production, build robust inference endpoints, optimize performance, and ensure models scale.
Build production-grade inference pipelines – design, deploy, and maintain ML services that handle real-time video processing; debug complex issues, optimize latency, and ensure 99.9% uptime.
Engineer video data workflows – build scalable preprocessing pipelines using serverless GPU infrastructure (RunPod, etc.) to transform raw video and audio into model-ready formats.
Architect cloud-native ML systems – leverage AWS (S3, DynamoDB, Lambda, ECS) and Kubernetes to build resilient, scalable data and inference infrastructure that can handle terabytes of video data.
Automate data annotation at scale – build and maintain labeling pipelines using AWS Ground Truth and Mechanical Turk.
Collaborate across teams – work closely with research and product teams to align model requirements and user needs.

2+ years of ML engineering, data engineering, or relevant experience
Experience building video/audio data processing pipelines using serverless GPU infrastructure like RunPod or similar providers.
Familiarity with machine learning and deep learning frameworks (PyTorch, TensorFlow)
Experience deploying ML models to inference platforms like Baseten or similar providers
Track record of adapting to new domains and using ML to improve products
Experience with AWS services (S3, DynamoDB) and containerization tools like Docker and Kubernetes
Languages: Python, SQL
Passionate about video AI, multimodal models, or conversational AI