About the role
We are looking for a creative technologist who understands both generative models and visual storytelling. The right person can move from experimentation to dependable workflows with measurable quality, repeatability, safety and cost control. Permanent role scope: This is a core technology position supporting multiple digital products. The person will own work through delivery, production operation, continuous enhancements and future versions, and may move across projects as priorities evolve.
Responsibilities:
- Design reusable prompt templates, shot structures, style systems and negative-prompt strategies for image and video generation.
- Build automated workflows for text-to-image, image-to-image, text-to-video, image-to-video, editing and variation generation.
- Integrate and evaluate multiple generative-media APIs and open-source models using secure, versioned interfaces.
- Create controls for composition, motion, camera behaviour, character or subject consistency, brand style and output format.
- Develop prompt chaining and multimodal workflows that convert briefs, references or storyboards into production-ready assets.
- Use Python and media-processing tools to automate generation, post-processing, transcoding, validation and asset management.
- Build evaluation rubrics and test sets for visual quality, prompt adherence, temporal consistency, safety, latency and cost.
- Document prompts, seeds, settings, model versions and source references so successful results can be reproduced.
- Apply copyright, likeness, consent, content-safety and data-handling controls to generated-media workflows.
- Collaborate with AI/ML, design, frontend, backend and QA teams to convert creative experiments into stable product capabilities.
Requirements:
- Strong hands-on prompt engineering for image and video generation, editing and controlled variation.
- Practical understanding of diffusion, transformer-based generative models, conditioning and multimodal workflows.
- Python scripting and experience integrating model APIs or inference pipelines.
- Working knowledge of visual composition, cinematography, shot planning, motion, pacing, aspect ratios and delivery formats.
- Ability to systematically compare outputs and improve quality through structured experimentation rather than subjective prompting alone.
- Experience with media processing and automation using tools such as FFmpeg, OpenCV or comparable libraries.
- Understanding of generative-media safety, copyright, consent, provenance and responsible-use requirements.
- A portfolio demonstrating high-quality, repeatable AI-generated image or video work and clear personal contribution.
Preferred experience:
- ComfyUI, Hugging Face Diffusers, ControlNet, LoRA, fine-tuning or custom inference workflows.
- Experience with major commercial image or video generation platforms and their APIs.
- Video editing, motion graphics, sound design, colour workflows or creative-production tools.
- Multimodal LLMs for brief interpretation, storyboard generation, prompt expansion or automated quality review.
- Experience building template-driven creative systems used by non-technical users.