Generative AI Engineer

(Image-to-Video Systems)

Responsibilities

Develop and maintain image-to-video generation pipelines.

Work directly with diffusion-based and transformer-based video models.

Design structured prompt formats for consistent and controllable outputs.

Implement systems to manage temporal consistency across frames.

Integrate generation pipelines with backend services.

Optimize pipelines for cost, performance, and output stability.

Evaluate and iterate on output quality using objective and subjective metrics.

Required Technical Experience

Mandatory

Hands-on experience with image-to-video or video generation models

Strong understanding of prompt engineering as a system, not one-off prompts

Proficiency in Python

Experience orchestrating multi-step AI pipelines

Familiarity with GPU-based inference workflows

Preferred

Experience with LoRA or fine-tuning generative models

Familiarity with tools for maintaining visual consistency

Experience working with multi-modal generative systems

Understanding of frame-level control and motion dynamics

This Role Is a Fit If You

Have built generative pipelines beyond demos

Understand model limitations and failure modes

Can reason about quality, consistency, and trade-offs

Generative AI Engineer