Visual Generation & Multimodal Evaluation Machine Learning Engineer Graduate (AML-Ark-US) - 2027 Start

Careers; San Jose (CA); On-site; USD 90,000 - 130,000; Full time

Job summary

ByteDance is seeking a graduate Machine Learning Engineer to build evaluation systems for image and video models and their agents. You will design reproducible human evaluation protocols and develop automated metrics to drive robust model assessments.

You will also design video generation agents that orchestrate multi-step creative workflows and create large-scale visual data pipelines to feed model improvements and product enhancements.

Qualifications

• Build evaluation systems for image and video models/agents, covering generation quality, instruction following, multimodal understanding, and safety.
• Develop automated metrics and model-based evaluators, and design reproducible human evaluation protocols.
• Design and develop video generation/debugging agents that orchestrate multi-step creative workflows.
• Build large-scale image and video data pipelines, and turn evaluation findings into model and product improvements.

Responsibilities

• Build evaluation systems for image and video models/agents, covering generation quality, instruction following, multimodal understanding, and safety
• Develop automated metrics and model-based evaluators, and design reproducible human evaluation protocols
• Design and develop video generation/debugging agents that orchestrate multi-step creative workflows
• Build large-scale image and video data pipelines, and turn evaluation findings into model and product improvements

Skills

Evaluation systems

Automated metrics

Human evaluation protocols

Video generation agents

Job description

ByteDance develops technology products and machine learning platforms that support creative expression and content experiences. The graduate Machine Learning Engineer will build evaluation systems and automated metrics for image and video models and agents, design reproducible human evaluation protocols, develop video-generation agents, and create large-scale visual data pipelines to improve models and products.

Responsibilities

• Build evaluation systems for image and video models/agents, covering generation quality, instruction following, multimodal understanding, and safety
• Develop automated metrics and model-based evaluators, and design reproducible human evaluation protocols
• Design and develop video generation/debugging agents that orchestrate multi-step creative workflows
• Build large-scale image and video data pipelines, and turn evaluation findings into model and product improvements