Generative AI Video Engineer

We are building a permanent AI persona for creative and video content generations. We require an engineer to establish an end-to-end pipeline that ensures 100% visual and vocal consistency across cinematic brand videos.

Responsibilities

• Identity Fixation: Design and train high-fidelity LoRA (Low-Rank Adaptation) models using Flux.1 Pro to "lock" a characters identity.
• Video Orchestration: Manage production pipelines using Seedance 2.0 (for lip-sync) and Kling 3.0 / Sora 2 (for cinematic action).
• Multimodal Sync: Integrate ElevenLabs v3 audio with generative video, ensuring perfect temporal alignment and micro-expression accuracy.
• Upscaling & Post-Process: Implement 4K upscaling workflows and temporal consistency checks (e.g., using Topaz AI or custom ComfyUI workflows).
  

Required Technical Skills:

• Models: Flux.1, SDXL, Kling 3.0, Sora 2, Seedance 2.0.
• Frameworks: ComfyUI (Expert level), AnimateDiff, IP-Adapter-FaceID.
• Training: Experience in training LoRAs for character consistency.
• Programming: Python (PyTorch), API integration for ElevenLabs and Video providers.

What we're looking for:

• A portfolio demonstrating Character Locking (the same person appearing in different scenes without morphing).
• A "Directors eye" for cinematic qualityunderstanding lighting, composition, and professional pacing.
• Experience with LivePortrait or similar facial expression transfer tools.